Transport Layer 3-1
Chapter 3Transport Layer
A note on the use of these ppt slidesThe notes used in this course are substantially based on powerpoint slides developed and copyrighted by JF Kurose and KW Ross 1996-2007
Computer Networking A Top Down Approach 4th edition Jim Kurose Keith RossAddison-Wesley July 2007
Transport Layer 3-2
Chapter 3 Transport LayerOur goals
Understand principles behind transport layer services
MultiplexingdemultiplexingReliable data transferFlow controlCongestion control
Learn about transport layer protocols in the Internet
UDP connectionless transportTCP connection-oriented transportTCP congestion control
Transport Layer 3-3
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-4
Transport Services and ProtocolsProvide logical communicationbetween app processes running on different hostsTransport protocols run in end systems
Send side breaks app messages into segments passes to network layerRcv side reassembles segments into messages passes to app layer
More than one transport protocol available to apps
Internet TCP and UDP
applicationtransportnetworkdata linkphysical
applicationtransportnetworkdata linkphysical
networkdata linkphysical
networkdata linkphysical
networkdata linkphysical
networkdata linkphysicalnetwork
data linkphysical
logical end-end transport
Transport Layer 3-5
Transport vs Network Layer
Network layerLogical communication between hosts
Transport layerLogical communication between processesRelies on enhances network layer services
Transport Layer 3-6
Internet Transport-Layer Protocols
Reliable in-order delivery (TCP)
Congestion control Flow controlConnection setup
Unreliable unordered delivery UDP
No-frills extension of ldquobest-effortrdquo IP
Services not available Delay guaranteesBandwidth guarantees
applicationtransportnetworkdata linkphysical
applicationtransportnetworkdata linkphysical
networkdata linkphysical
networkdata linkphysical
networkdata linkphysical
networkdata linkphysicalnetwork
data linkphysical
logical end-end transport
Transport Layer 3-7
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-8
MultiplexingDemultiplexing
host 1
= process= socket
1 host 1 or more processes1 process 1 or more socketsTransport layer interacts with socket
bull Demultiplexing at rcv hostbull Multiplexing at send host
application
transport
network
link
physical
P1 application
transport
network
link
physical
application
transport
network
link
physical
P2P3 P4P1
host 1 host 2 host 3
Transport Layer 3-9
How Demultiplexing WorksHost receives IP datagrams
Each datagram has source IP address destination IP addressEach datagram carries 1 transport-layer segmentEach segment has source destination port number (recall well-known port numbers for specific applications)
Host uses IP addresses amp port numbers to direct segment to appropriate socket
source port dest port
32 bits
applicationdata
(message)
other header fields
TCPUDP segment format
Transport Layer 3-10
Connectionless Demultiplexing
Create sockets with port numbers
DatagramSocket mySocket1 = new DatagramSocket(12534)
DatagramSocket mySocket2 = new DatagramSocket(12535)
UDP socket identified by two-tuple
(dest IP address dest port number)
When host receives UDP segment
Checks destination port number in segmentDirects UDP segment to socket with that port number
IP datagrams with different source IP addresses andor source port numbers directed to same socket
Transport Layer 3-11
Connection-Oriented Demux
TCP socket identified by 4-tuple
Source IP addressSource port numberDest IP addressDest port number
Recv host uses all four values to direct segment to appropriate socket
Server host may support many simultaneous TCP sockets
Each socket identified by its own 4-tuple
Web servers have different sockets for each connecting client
Non-persistent HTTP will have different socket for each request
Transport Layer 3-12
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-13
UDP User Datagram Protocol [RFC 768]
ldquoNo frillsrdquo ldquobare bonesrdquoInternet transport protocolldquoBest effortrdquo service UDP segments may be
LostDelivered out of order to app
ConnectionlessNo handshaking between UDP sender receiverEach UDP segment handled independently of others
Why is there a UDPNo connection establishment (which can add delay)Simple no connection state at sender receiverSmall segment headerNo congestion control UDP can blast away as fast as desired
Transport Layer 3-14
UDP moreOften used for streaming multimedia apps
Loss tolerantRate sensitive
Other UDP usesDNSSNMP
Reliable transfer over UDP add reliability at application layer
Application-specific error recovery
source port dest port
32 bits
Applicationdata
(message)
UDP segment format
length checksumLength in
bytes of UDPsegmentincluding
header
Transport Layer 3-15
UDP Checksum
SenderTreat segment contents as sequence of 16-bit integersChecksum addition (1rsquos complement sum) of segment contentsSender puts checksum value into UDP checksum field
ReceiverCompute checksum of received segmentCheck if computed checksum equals checksum field value
NO - error detectedYES - no error detected But maybe errors nonetheless More later
Goal detect ldquoerrorsrdquo (eg flipped bits) in transmitted segment
Transport Layer 3-16
Internet Checksum ExampleNote
When adding numbers a carryout from the most significant bit needs to be added to the result
Example add two 16-bit integers
wraparound
sumchecksum
1 1 1 1 0 0 1 1 0 0 1 1 0 0 1 1 01 1 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 0 1 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 1 0 01 0 1 0 0 0 1 0 0 0 1 0 0 0 0 1 1
Transport Layer 3-17
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-18
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Transport Layer 3-19
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-20
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-21
Reliable Data Transfer Getting Started
sendside
receiveside
rdt_send() called from above (eg by app) Passed data to
deliver to receiver upper layer
udt_send() called by rdtto transfer packet over
unreliable channel to receiver
rdt_rcv() called when packet arrives on rcv-side of channel
deliver_data() called by rdt to deliver data to upper
Transport Layer 3-22
Reliable Data Transfer Getting StartedWersquoll
Incrementally develop sender receiver sides of reliable data transfer protocol (rdt)Consider only unidirectional data transfer
But control info will flow on both directionsUse finite state machines (FSM) to specify sender receiver
state1
state2
event causing state transitionactions taken on state transition
state when in this ldquostaterdquo next state
uniquely determined by next event
eventactions
Transport Layer 3-23
Summary of the Protocols
Rdt10 all packets arrive correctlyRdt20 all packets arrive with possible errors only in data packets and introducing ACK and NAK (no error)Rdt21 corrupted ACKsNAKsRdt22 similar to rdt21 remove NAKsRdt30 Allows packets to be lost and errors
Transport Layer 3-24
Rdt10 Reliable Transfer over a Reliable Channel
Underlying channel perfectly reliableNo bit errorsNo loss of packets
Separate FSMs for sender receiverSender sends data into underlying channelReceiver read data from underlying channel
packet = make_pkt(data)udt_send(packet)
Wait for call from above
rdt_send(data)
sender
extract (packetdata)deliver_data(data)
Wait for call from
below
rdt_rcv(packet)
receiver
Transport Layer 3-25
Rdt20 Channel with Bit Errors
Underlying channel may flip bits in packetChecksum to detect bit errors
The question how to recover from errorsAcknowledgements (ACKs) receiver explicitly tells sender that pkt received OKNegative acknowledgements (NAKs) receiver explicitly tells sender that pkt had errorsSender retransmits pkt on receipt of NAK
New mechanisms in rdt20 (beyond rdt10)Error detectionReceiver feedback control msgs (ACKNAK) rcvr-gtsender
Transport Layer 3-26
Rdt20 FSM specification
Wait for call from above
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
sender
receiversnkpkt = make_pkt(data checksum)udt_send(sndpkt)
rdt_send(data)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)Λ
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-2
Chapter 3 Transport LayerOur goals
Understand principles behind transport layer services
MultiplexingdemultiplexingReliable data transferFlow controlCongestion control
Learn about transport layer protocols in the Internet
UDP connectionless transportTCP connection-oriented transportTCP congestion control
Transport Layer 3-3
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-4
Transport Services and ProtocolsProvide logical communicationbetween app processes running on different hostsTransport protocols run in end systems
Send side breaks app messages into segments passes to network layerRcv side reassembles segments into messages passes to app layer
More than one transport protocol available to apps
Internet TCP and UDP
applicationtransportnetworkdata linkphysical
applicationtransportnetworkdata linkphysical
networkdata linkphysical
networkdata linkphysical
networkdata linkphysical
networkdata linkphysicalnetwork
data linkphysical
logical end-end transport
Transport Layer 3-5
Transport vs Network Layer
Network layerLogical communication between hosts
Transport layerLogical communication between processesRelies on enhances network layer services
Transport Layer 3-6
Internet Transport-Layer Protocols
Reliable in-order delivery (TCP)
Congestion control Flow controlConnection setup
Unreliable unordered delivery UDP
No-frills extension of ldquobest-effortrdquo IP
Services not available Delay guaranteesBandwidth guarantees
applicationtransportnetworkdata linkphysical
applicationtransportnetworkdata linkphysical
networkdata linkphysical
networkdata linkphysical
networkdata linkphysical
networkdata linkphysicalnetwork
data linkphysical
logical end-end transport
Transport Layer 3-7
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-8
MultiplexingDemultiplexing
host 1
= process= socket
1 host 1 or more processes1 process 1 or more socketsTransport layer interacts with socket
bull Demultiplexing at rcv hostbull Multiplexing at send host
application
transport
network
link
physical
P1 application
transport
network
link
physical
application
transport
network
link
physical
P2P3 P4P1
host 1 host 2 host 3
Transport Layer 3-9
How Demultiplexing WorksHost receives IP datagrams
Each datagram has source IP address destination IP addressEach datagram carries 1 transport-layer segmentEach segment has source destination port number (recall well-known port numbers for specific applications)
Host uses IP addresses amp port numbers to direct segment to appropriate socket
source port dest port
32 bits
applicationdata
(message)
other header fields
TCPUDP segment format
Transport Layer 3-10
Connectionless Demultiplexing
Create sockets with port numbers
DatagramSocket mySocket1 = new DatagramSocket(12534)
DatagramSocket mySocket2 = new DatagramSocket(12535)
UDP socket identified by two-tuple
(dest IP address dest port number)
When host receives UDP segment
Checks destination port number in segmentDirects UDP segment to socket with that port number
IP datagrams with different source IP addresses andor source port numbers directed to same socket
Transport Layer 3-11
Connection-Oriented Demux
TCP socket identified by 4-tuple
Source IP addressSource port numberDest IP addressDest port number
Recv host uses all four values to direct segment to appropriate socket
Server host may support many simultaneous TCP sockets
Each socket identified by its own 4-tuple
Web servers have different sockets for each connecting client
Non-persistent HTTP will have different socket for each request
Transport Layer 3-12
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-13
UDP User Datagram Protocol [RFC 768]
ldquoNo frillsrdquo ldquobare bonesrdquoInternet transport protocolldquoBest effortrdquo service UDP segments may be
LostDelivered out of order to app
ConnectionlessNo handshaking between UDP sender receiverEach UDP segment handled independently of others
Why is there a UDPNo connection establishment (which can add delay)Simple no connection state at sender receiverSmall segment headerNo congestion control UDP can blast away as fast as desired
Transport Layer 3-14
UDP moreOften used for streaming multimedia apps
Loss tolerantRate sensitive
Other UDP usesDNSSNMP
Reliable transfer over UDP add reliability at application layer
Application-specific error recovery
source port dest port
32 bits
Applicationdata
(message)
UDP segment format
length checksumLength in
bytes of UDPsegmentincluding
header
Transport Layer 3-15
UDP Checksum
SenderTreat segment contents as sequence of 16-bit integersChecksum addition (1rsquos complement sum) of segment contentsSender puts checksum value into UDP checksum field
ReceiverCompute checksum of received segmentCheck if computed checksum equals checksum field value
NO - error detectedYES - no error detected But maybe errors nonetheless More later
Goal detect ldquoerrorsrdquo (eg flipped bits) in transmitted segment
Transport Layer 3-16
Internet Checksum ExampleNote
When adding numbers a carryout from the most significant bit needs to be added to the result
Example add two 16-bit integers
wraparound
sumchecksum
1 1 1 1 0 0 1 1 0 0 1 1 0 0 1 1 01 1 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 0 1 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 1 0 01 0 1 0 0 0 1 0 0 0 1 0 0 0 0 1 1
Transport Layer 3-17
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-18
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Transport Layer 3-19
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-20
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-21
Reliable Data Transfer Getting Started
sendside
receiveside
rdt_send() called from above (eg by app) Passed data to
deliver to receiver upper layer
udt_send() called by rdtto transfer packet over
unreliable channel to receiver
rdt_rcv() called when packet arrives on rcv-side of channel
deliver_data() called by rdt to deliver data to upper
Transport Layer 3-22
Reliable Data Transfer Getting StartedWersquoll
Incrementally develop sender receiver sides of reliable data transfer protocol (rdt)Consider only unidirectional data transfer
But control info will flow on both directionsUse finite state machines (FSM) to specify sender receiver
state1
state2
event causing state transitionactions taken on state transition
state when in this ldquostaterdquo next state
uniquely determined by next event
eventactions
Transport Layer 3-23
Summary of the Protocols
Rdt10 all packets arrive correctlyRdt20 all packets arrive with possible errors only in data packets and introducing ACK and NAK (no error)Rdt21 corrupted ACKsNAKsRdt22 similar to rdt21 remove NAKsRdt30 Allows packets to be lost and errors
Transport Layer 3-24
Rdt10 Reliable Transfer over a Reliable Channel
Underlying channel perfectly reliableNo bit errorsNo loss of packets
Separate FSMs for sender receiverSender sends data into underlying channelReceiver read data from underlying channel
packet = make_pkt(data)udt_send(packet)
Wait for call from above
rdt_send(data)
sender
extract (packetdata)deliver_data(data)
Wait for call from
below
rdt_rcv(packet)
receiver
Transport Layer 3-25
Rdt20 Channel with Bit Errors
Underlying channel may flip bits in packetChecksum to detect bit errors
The question how to recover from errorsAcknowledgements (ACKs) receiver explicitly tells sender that pkt received OKNegative acknowledgements (NAKs) receiver explicitly tells sender that pkt had errorsSender retransmits pkt on receipt of NAK
New mechanisms in rdt20 (beyond rdt10)Error detectionReceiver feedback control msgs (ACKNAK) rcvr-gtsender
Transport Layer 3-26
Rdt20 FSM specification
Wait for call from above
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
sender
receiversnkpkt = make_pkt(data checksum)udt_send(sndpkt)
rdt_send(data)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)Λ
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-3
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-4
Transport Services and ProtocolsProvide logical communicationbetween app processes running on different hostsTransport protocols run in end systems
Send side breaks app messages into segments passes to network layerRcv side reassembles segments into messages passes to app layer
More than one transport protocol available to apps
Internet TCP and UDP
applicationtransportnetworkdata linkphysical
applicationtransportnetworkdata linkphysical
networkdata linkphysical
networkdata linkphysical
networkdata linkphysical
networkdata linkphysicalnetwork
data linkphysical
logical end-end transport
Transport Layer 3-5
Transport vs Network Layer
Network layerLogical communication between hosts
Transport layerLogical communication between processesRelies on enhances network layer services
Transport Layer 3-6
Internet Transport-Layer Protocols
Reliable in-order delivery (TCP)
Congestion control Flow controlConnection setup
Unreliable unordered delivery UDP
No-frills extension of ldquobest-effortrdquo IP
Services not available Delay guaranteesBandwidth guarantees
applicationtransportnetworkdata linkphysical
applicationtransportnetworkdata linkphysical
networkdata linkphysical
networkdata linkphysical
networkdata linkphysical
networkdata linkphysicalnetwork
data linkphysical
logical end-end transport
Transport Layer 3-7
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-8
MultiplexingDemultiplexing
host 1
= process= socket
1 host 1 or more processes1 process 1 or more socketsTransport layer interacts with socket
bull Demultiplexing at rcv hostbull Multiplexing at send host
application
transport
network
link
physical
P1 application
transport
network
link
physical
application
transport
network
link
physical
P2P3 P4P1
host 1 host 2 host 3
Transport Layer 3-9
How Demultiplexing WorksHost receives IP datagrams
Each datagram has source IP address destination IP addressEach datagram carries 1 transport-layer segmentEach segment has source destination port number (recall well-known port numbers for specific applications)
Host uses IP addresses amp port numbers to direct segment to appropriate socket
source port dest port
32 bits
applicationdata
(message)
other header fields
TCPUDP segment format
Transport Layer 3-10
Connectionless Demultiplexing
Create sockets with port numbers
DatagramSocket mySocket1 = new DatagramSocket(12534)
DatagramSocket mySocket2 = new DatagramSocket(12535)
UDP socket identified by two-tuple
(dest IP address dest port number)
When host receives UDP segment
Checks destination port number in segmentDirects UDP segment to socket with that port number
IP datagrams with different source IP addresses andor source port numbers directed to same socket
Transport Layer 3-11
Connection-Oriented Demux
TCP socket identified by 4-tuple
Source IP addressSource port numberDest IP addressDest port number
Recv host uses all four values to direct segment to appropriate socket
Server host may support many simultaneous TCP sockets
Each socket identified by its own 4-tuple
Web servers have different sockets for each connecting client
Non-persistent HTTP will have different socket for each request
Transport Layer 3-12
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-13
UDP User Datagram Protocol [RFC 768]
ldquoNo frillsrdquo ldquobare bonesrdquoInternet transport protocolldquoBest effortrdquo service UDP segments may be
LostDelivered out of order to app
ConnectionlessNo handshaking between UDP sender receiverEach UDP segment handled independently of others
Why is there a UDPNo connection establishment (which can add delay)Simple no connection state at sender receiverSmall segment headerNo congestion control UDP can blast away as fast as desired
Transport Layer 3-14
UDP moreOften used for streaming multimedia apps
Loss tolerantRate sensitive
Other UDP usesDNSSNMP
Reliable transfer over UDP add reliability at application layer
Application-specific error recovery
source port dest port
32 bits
Applicationdata
(message)
UDP segment format
length checksumLength in
bytes of UDPsegmentincluding
header
Transport Layer 3-15
UDP Checksum
SenderTreat segment contents as sequence of 16-bit integersChecksum addition (1rsquos complement sum) of segment contentsSender puts checksum value into UDP checksum field
ReceiverCompute checksum of received segmentCheck if computed checksum equals checksum field value
NO - error detectedYES - no error detected But maybe errors nonetheless More later
Goal detect ldquoerrorsrdquo (eg flipped bits) in transmitted segment
Transport Layer 3-16
Internet Checksum ExampleNote
When adding numbers a carryout from the most significant bit needs to be added to the result
Example add two 16-bit integers
wraparound
sumchecksum
1 1 1 1 0 0 1 1 0 0 1 1 0 0 1 1 01 1 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 0 1 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 1 0 01 0 1 0 0 0 1 0 0 0 1 0 0 0 0 1 1
Transport Layer 3-17
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-18
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Transport Layer 3-19
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-20
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-21
Reliable Data Transfer Getting Started
sendside
receiveside
rdt_send() called from above (eg by app) Passed data to
deliver to receiver upper layer
udt_send() called by rdtto transfer packet over
unreliable channel to receiver
rdt_rcv() called when packet arrives on rcv-side of channel
deliver_data() called by rdt to deliver data to upper
Transport Layer 3-22
Reliable Data Transfer Getting StartedWersquoll
Incrementally develop sender receiver sides of reliable data transfer protocol (rdt)Consider only unidirectional data transfer
But control info will flow on both directionsUse finite state machines (FSM) to specify sender receiver
state1
state2
event causing state transitionactions taken on state transition
state when in this ldquostaterdquo next state
uniquely determined by next event
eventactions
Transport Layer 3-23
Summary of the Protocols
Rdt10 all packets arrive correctlyRdt20 all packets arrive with possible errors only in data packets and introducing ACK and NAK (no error)Rdt21 corrupted ACKsNAKsRdt22 similar to rdt21 remove NAKsRdt30 Allows packets to be lost and errors
Transport Layer 3-24
Rdt10 Reliable Transfer over a Reliable Channel
Underlying channel perfectly reliableNo bit errorsNo loss of packets
Separate FSMs for sender receiverSender sends data into underlying channelReceiver read data from underlying channel
packet = make_pkt(data)udt_send(packet)
Wait for call from above
rdt_send(data)
sender
extract (packetdata)deliver_data(data)
Wait for call from
below
rdt_rcv(packet)
receiver
Transport Layer 3-25
Rdt20 Channel with Bit Errors
Underlying channel may flip bits in packetChecksum to detect bit errors
The question how to recover from errorsAcknowledgements (ACKs) receiver explicitly tells sender that pkt received OKNegative acknowledgements (NAKs) receiver explicitly tells sender that pkt had errorsSender retransmits pkt on receipt of NAK
New mechanisms in rdt20 (beyond rdt10)Error detectionReceiver feedback control msgs (ACKNAK) rcvr-gtsender
Transport Layer 3-26
Rdt20 FSM specification
Wait for call from above
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
sender
receiversnkpkt = make_pkt(data checksum)udt_send(sndpkt)
rdt_send(data)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)Λ
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-4
Transport Services and ProtocolsProvide logical communicationbetween app processes running on different hostsTransport protocols run in end systems
Send side breaks app messages into segments passes to network layerRcv side reassembles segments into messages passes to app layer
More than one transport protocol available to apps
Internet TCP and UDP
applicationtransportnetworkdata linkphysical
applicationtransportnetworkdata linkphysical
networkdata linkphysical
networkdata linkphysical
networkdata linkphysical
networkdata linkphysicalnetwork
data linkphysical
logical end-end transport
Transport Layer 3-5
Transport vs Network Layer
Network layerLogical communication between hosts
Transport layerLogical communication between processesRelies on enhances network layer services
Transport Layer 3-6
Internet Transport-Layer Protocols
Reliable in-order delivery (TCP)
Congestion control Flow controlConnection setup
Unreliable unordered delivery UDP
No-frills extension of ldquobest-effortrdquo IP
Services not available Delay guaranteesBandwidth guarantees
applicationtransportnetworkdata linkphysical
applicationtransportnetworkdata linkphysical
networkdata linkphysical
networkdata linkphysical
networkdata linkphysical
networkdata linkphysicalnetwork
data linkphysical
logical end-end transport
Transport Layer 3-7
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-8
MultiplexingDemultiplexing
host 1
= process= socket
1 host 1 or more processes1 process 1 or more socketsTransport layer interacts with socket
bull Demultiplexing at rcv hostbull Multiplexing at send host
application
transport
network
link
physical
P1 application
transport
network
link
physical
application
transport
network
link
physical
P2P3 P4P1
host 1 host 2 host 3
Transport Layer 3-9
How Demultiplexing WorksHost receives IP datagrams
Each datagram has source IP address destination IP addressEach datagram carries 1 transport-layer segmentEach segment has source destination port number (recall well-known port numbers for specific applications)
Host uses IP addresses amp port numbers to direct segment to appropriate socket
source port dest port
32 bits
applicationdata
(message)
other header fields
TCPUDP segment format
Transport Layer 3-10
Connectionless Demultiplexing
Create sockets with port numbers
DatagramSocket mySocket1 = new DatagramSocket(12534)
DatagramSocket mySocket2 = new DatagramSocket(12535)
UDP socket identified by two-tuple
(dest IP address dest port number)
When host receives UDP segment
Checks destination port number in segmentDirects UDP segment to socket with that port number
IP datagrams with different source IP addresses andor source port numbers directed to same socket
Transport Layer 3-11
Connection-Oriented Demux
TCP socket identified by 4-tuple
Source IP addressSource port numberDest IP addressDest port number
Recv host uses all four values to direct segment to appropriate socket
Server host may support many simultaneous TCP sockets
Each socket identified by its own 4-tuple
Web servers have different sockets for each connecting client
Non-persistent HTTP will have different socket for each request
Transport Layer 3-12
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-13
UDP User Datagram Protocol [RFC 768]
ldquoNo frillsrdquo ldquobare bonesrdquoInternet transport protocolldquoBest effortrdquo service UDP segments may be
LostDelivered out of order to app
ConnectionlessNo handshaking between UDP sender receiverEach UDP segment handled independently of others
Why is there a UDPNo connection establishment (which can add delay)Simple no connection state at sender receiverSmall segment headerNo congestion control UDP can blast away as fast as desired
Transport Layer 3-14
UDP moreOften used for streaming multimedia apps
Loss tolerantRate sensitive
Other UDP usesDNSSNMP
Reliable transfer over UDP add reliability at application layer
Application-specific error recovery
source port dest port
32 bits
Applicationdata
(message)
UDP segment format
length checksumLength in
bytes of UDPsegmentincluding
header
Transport Layer 3-15
UDP Checksum
SenderTreat segment contents as sequence of 16-bit integersChecksum addition (1rsquos complement sum) of segment contentsSender puts checksum value into UDP checksum field
ReceiverCompute checksum of received segmentCheck if computed checksum equals checksum field value
NO - error detectedYES - no error detected But maybe errors nonetheless More later
Goal detect ldquoerrorsrdquo (eg flipped bits) in transmitted segment
Transport Layer 3-16
Internet Checksum ExampleNote
When adding numbers a carryout from the most significant bit needs to be added to the result
Example add two 16-bit integers
wraparound
sumchecksum
1 1 1 1 0 0 1 1 0 0 1 1 0 0 1 1 01 1 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 0 1 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 1 0 01 0 1 0 0 0 1 0 0 0 1 0 0 0 0 1 1
Transport Layer 3-17
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-18
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Transport Layer 3-19
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-20
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-21
Reliable Data Transfer Getting Started
sendside
receiveside
rdt_send() called from above (eg by app) Passed data to
deliver to receiver upper layer
udt_send() called by rdtto transfer packet over
unreliable channel to receiver
rdt_rcv() called when packet arrives on rcv-side of channel
deliver_data() called by rdt to deliver data to upper
Transport Layer 3-22
Reliable Data Transfer Getting StartedWersquoll
Incrementally develop sender receiver sides of reliable data transfer protocol (rdt)Consider only unidirectional data transfer
But control info will flow on both directionsUse finite state machines (FSM) to specify sender receiver
state1
state2
event causing state transitionactions taken on state transition
state when in this ldquostaterdquo next state
uniquely determined by next event
eventactions
Transport Layer 3-23
Summary of the Protocols
Rdt10 all packets arrive correctlyRdt20 all packets arrive with possible errors only in data packets and introducing ACK and NAK (no error)Rdt21 corrupted ACKsNAKsRdt22 similar to rdt21 remove NAKsRdt30 Allows packets to be lost and errors
Transport Layer 3-24
Rdt10 Reliable Transfer over a Reliable Channel
Underlying channel perfectly reliableNo bit errorsNo loss of packets
Separate FSMs for sender receiverSender sends data into underlying channelReceiver read data from underlying channel
packet = make_pkt(data)udt_send(packet)
Wait for call from above
rdt_send(data)
sender
extract (packetdata)deliver_data(data)
Wait for call from
below
rdt_rcv(packet)
receiver
Transport Layer 3-25
Rdt20 Channel with Bit Errors
Underlying channel may flip bits in packetChecksum to detect bit errors
The question how to recover from errorsAcknowledgements (ACKs) receiver explicitly tells sender that pkt received OKNegative acknowledgements (NAKs) receiver explicitly tells sender that pkt had errorsSender retransmits pkt on receipt of NAK
New mechanisms in rdt20 (beyond rdt10)Error detectionReceiver feedback control msgs (ACKNAK) rcvr-gtsender
Transport Layer 3-26
Rdt20 FSM specification
Wait for call from above
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
sender
receiversnkpkt = make_pkt(data checksum)udt_send(sndpkt)
rdt_send(data)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)Λ
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-5
Transport vs Network Layer
Network layerLogical communication between hosts
Transport layerLogical communication between processesRelies on enhances network layer services
Transport Layer 3-6
Internet Transport-Layer Protocols
Reliable in-order delivery (TCP)
Congestion control Flow controlConnection setup
Unreliable unordered delivery UDP
No-frills extension of ldquobest-effortrdquo IP
Services not available Delay guaranteesBandwidth guarantees
applicationtransportnetworkdata linkphysical
applicationtransportnetworkdata linkphysical
networkdata linkphysical
networkdata linkphysical
networkdata linkphysical
networkdata linkphysicalnetwork
data linkphysical
logical end-end transport
Transport Layer 3-7
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-8
MultiplexingDemultiplexing
host 1
= process= socket
1 host 1 or more processes1 process 1 or more socketsTransport layer interacts with socket
bull Demultiplexing at rcv hostbull Multiplexing at send host
application
transport
network
link
physical
P1 application
transport
network
link
physical
application
transport
network
link
physical
P2P3 P4P1
host 1 host 2 host 3
Transport Layer 3-9
How Demultiplexing WorksHost receives IP datagrams
Each datagram has source IP address destination IP addressEach datagram carries 1 transport-layer segmentEach segment has source destination port number (recall well-known port numbers for specific applications)
Host uses IP addresses amp port numbers to direct segment to appropriate socket
source port dest port
32 bits
applicationdata
(message)
other header fields
TCPUDP segment format
Transport Layer 3-10
Connectionless Demultiplexing
Create sockets with port numbers
DatagramSocket mySocket1 = new DatagramSocket(12534)
DatagramSocket mySocket2 = new DatagramSocket(12535)
UDP socket identified by two-tuple
(dest IP address dest port number)
When host receives UDP segment
Checks destination port number in segmentDirects UDP segment to socket with that port number
IP datagrams with different source IP addresses andor source port numbers directed to same socket
Transport Layer 3-11
Connection-Oriented Demux
TCP socket identified by 4-tuple
Source IP addressSource port numberDest IP addressDest port number
Recv host uses all four values to direct segment to appropriate socket
Server host may support many simultaneous TCP sockets
Each socket identified by its own 4-tuple
Web servers have different sockets for each connecting client
Non-persistent HTTP will have different socket for each request
Transport Layer 3-12
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-13
UDP User Datagram Protocol [RFC 768]
ldquoNo frillsrdquo ldquobare bonesrdquoInternet transport protocolldquoBest effortrdquo service UDP segments may be
LostDelivered out of order to app
ConnectionlessNo handshaking between UDP sender receiverEach UDP segment handled independently of others
Why is there a UDPNo connection establishment (which can add delay)Simple no connection state at sender receiverSmall segment headerNo congestion control UDP can blast away as fast as desired
Transport Layer 3-14
UDP moreOften used for streaming multimedia apps
Loss tolerantRate sensitive
Other UDP usesDNSSNMP
Reliable transfer over UDP add reliability at application layer
Application-specific error recovery
source port dest port
32 bits
Applicationdata
(message)
UDP segment format
length checksumLength in
bytes of UDPsegmentincluding
header
Transport Layer 3-15
UDP Checksum
SenderTreat segment contents as sequence of 16-bit integersChecksum addition (1rsquos complement sum) of segment contentsSender puts checksum value into UDP checksum field
ReceiverCompute checksum of received segmentCheck if computed checksum equals checksum field value
NO - error detectedYES - no error detected But maybe errors nonetheless More later
Goal detect ldquoerrorsrdquo (eg flipped bits) in transmitted segment
Transport Layer 3-16
Internet Checksum ExampleNote
When adding numbers a carryout from the most significant bit needs to be added to the result
Example add two 16-bit integers
wraparound
sumchecksum
1 1 1 1 0 0 1 1 0 0 1 1 0 0 1 1 01 1 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 0 1 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 1 0 01 0 1 0 0 0 1 0 0 0 1 0 0 0 0 1 1
Transport Layer 3-17
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-18
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Transport Layer 3-19
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-20
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-21
Reliable Data Transfer Getting Started
sendside
receiveside
rdt_send() called from above (eg by app) Passed data to
deliver to receiver upper layer
udt_send() called by rdtto transfer packet over
unreliable channel to receiver
rdt_rcv() called when packet arrives on rcv-side of channel
deliver_data() called by rdt to deliver data to upper
Transport Layer 3-22
Reliable Data Transfer Getting StartedWersquoll
Incrementally develop sender receiver sides of reliable data transfer protocol (rdt)Consider only unidirectional data transfer
But control info will flow on both directionsUse finite state machines (FSM) to specify sender receiver
state1
state2
event causing state transitionactions taken on state transition
state when in this ldquostaterdquo next state
uniquely determined by next event
eventactions
Transport Layer 3-23
Summary of the Protocols
Rdt10 all packets arrive correctlyRdt20 all packets arrive with possible errors only in data packets and introducing ACK and NAK (no error)Rdt21 corrupted ACKsNAKsRdt22 similar to rdt21 remove NAKsRdt30 Allows packets to be lost and errors
Transport Layer 3-24
Rdt10 Reliable Transfer over a Reliable Channel
Underlying channel perfectly reliableNo bit errorsNo loss of packets
Separate FSMs for sender receiverSender sends data into underlying channelReceiver read data from underlying channel
packet = make_pkt(data)udt_send(packet)
Wait for call from above
rdt_send(data)
sender
extract (packetdata)deliver_data(data)
Wait for call from
below
rdt_rcv(packet)
receiver
Transport Layer 3-25
Rdt20 Channel with Bit Errors
Underlying channel may flip bits in packetChecksum to detect bit errors
The question how to recover from errorsAcknowledgements (ACKs) receiver explicitly tells sender that pkt received OKNegative acknowledgements (NAKs) receiver explicitly tells sender that pkt had errorsSender retransmits pkt on receipt of NAK
New mechanisms in rdt20 (beyond rdt10)Error detectionReceiver feedback control msgs (ACKNAK) rcvr-gtsender
Transport Layer 3-26
Rdt20 FSM specification
Wait for call from above
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
sender
receiversnkpkt = make_pkt(data checksum)udt_send(sndpkt)
rdt_send(data)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)Λ
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-6
Internet Transport-Layer Protocols
Reliable in-order delivery (TCP)
Congestion control Flow controlConnection setup
Unreliable unordered delivery UDP
No-frills extension of ldquobest-effortrdquo IP
Services not available Delay guaranteesBandwidth guarantees
applicationtransportnetworkdata linkphysical
applicationtransportnetworkdata linkphysical
networkdata linkphysical
networkdata linkphysical
networkdata linkphysical
networkdata linkphysicalnetwork
data linkphysical
logical end-end transport
Transport Layer 3-7
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-8
MultiplexingDemultiplexing
host 1
= process= socket
1 host 1 or more processes1 process 1 or more socketsTransport layer interacts with socket
bull Demultiplexing at rcv hostbull Multiplexing at send host
application
transport
network
link
physical
P1 application
transport
network
link
physical
application
transport
network
link
physical
P2P3 P4P1
host 1 host 2 host 3
Transport Layer 3-9
How Demultiplexing WorksHost receives IP datagrams
Each datagram has source IP address destination IP addressEach datagram carries 1 transport-layer segmentEach segment has source destination port number (recall well-known port numbers for specific applications)
Host uses IP addresses amp port numbers to direct segment to appropriate socket
source port dest port
32 bits
applicationdata
(message)
other header fields
TCPUDP segment format
Transport Layer 3-10
Connectionless Demultiplexing
Create sockets with port numbers
DatagramSocket mySocket1 = new DatagramSocket(12534)
DatagramSocket mySocket2 = new DatagramSocket(12535)
UDP socket identified by two-tuple
(dest IP address dest port number)
When host receives UDP segment
Checks destination port number in segmentDirects UDP segment to socket with that port number
IP datagrams with different source IP addresses andor source port numbers directed to same socket
Transport Layer 3-11
Connection-Oriented Demux
TCP socket identified by 4-tuple
Source IP addressSource port numberDest IP addressDest port number
Recv host uses all four values to direct segment to appropriate socket
Server host may support many simultaneous TCP sockets
Each socket identified by its own 4-tuple
Web servers have different sockets for each connecting client
Non-persistent HTTP will have different socket for each request
Transport Layer 3-12
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-13
UDP User Datagram Protocol [RFC 768]
ldquoNo frillsrdquo ldquobare bonesrdquoInternet transport protocolldquoBest effortrdquo service UDP segments may be
LostDelivered out of order to app
ConnectionlessNo handshaking between UDP sender receiverEach UDP segment handled independently of others
Why is there a UDPNo connection establishment (which can add delay)Simple no connection state at sender receiverSmall segment headerNo congestion control UDP can blast away as fast as desired
Transport Layer 3-14
UDP moreOften used for streaming multimedia apps
Loss tolerantRate sensitive
Other UDP usesDNSSNMP
Reliable transfer over UDP add reliability at application layer
Application-specific error recovery
source port dest port
32 bits
Applicationdata
(message)
UDP segment format
length checksumLength in
bytes of UDPsegmentincluding
header
Transport Layer 3-15
UDP Checksum
SenderTreat segment contents as sequence of 16-bit integersChecksum addition (1rsquos complement sum) of segment contentsSender puts checksum value into UDP checksum field
ReceiverCompute checksum of received segmentCheck if computed checksum equals checksum field value
NO - error detectedYES - no error detected But maybe errors nonetheless More later
Goal detect ldquoerrorsrdquo (eg flipped bits) in transmitted segment
Transport Layer 3-16
Internet Checksum ExampleNote
When adding numbers a carryout from the most significant bit needs to be added to the result
Example add two 16-bit integers
wraparound
sumchecksum
1 1 1 1 0 0 1 1 0 0 1 1 0 0 1 1 01 1 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 0 1 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 1 0 01 0 1 0 0 0 1 0 0 0 1 0 0 0 0 1 1
Transport Layer 3-17
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-18
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Transport Layer 3-19
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-20
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-21
Reliable Data Transfer Getting Started
sendside
receiveside
rdt_send() called from above (eg by app) Passed data to
deliver to receiver upper layer
udt_send() called by rdtto transfer packet over
unreliable channel to receiver
rdt_rcv() called when packet arrives on rcv-side of channel
deliver_data() called by rdt to deliver data to upper
Transport Layer 3-22
Reliable Data Transfer Getting StartedWersquoll
Incrementally develop sender receiver sides of reliable data transfer protocol (rdt)Consider only unidirectional data transfer
But control info will flow on both directionsUse finite state machines (FSM) to specify sender receiver
state1
state2
event causing state transitionactions taken on state transition
state when in this ldquostaterdquo next state
uniquely determined by next event
eventactions
Transport Layer 3-23
Summary of the Protocols
Rdt10 all packets arrive correctlyRdt20 all packets arrive with possible errors only in data packets and introducing ACK and NAK (no error)Rdt21 corrupted ACKsNAKsRdt22 similar to rdt21 remove NAKsRdt30 Allows packets to be lost and errors
Transport Layer 3-24
Rdt10 Reliable Transfer over a Reliable Channel
Underlying channel perfectly reliableNo bit errorsNo loss of packets
Separate FSMs for sender receiverSender sends data into underlying channelReceiver read data from underlying channel
packet = make_pkt(data)udt_send(packet)
Wait for call from above
rdt_send(data)
sender
extract (packetdata)deliver_data(data)
Wait for call from
below
rdt_rcv(packet)
receiver
Transport Layer 3-25
Rdt20 Channel with Bit Errors
Underlying channel may flip bits in packetChecksum to detect bit errors
The question how to recover from errorsAcknowledgements (ACKs) receiver explicitly tells sender that pkt received OKNegative acknowledgements (NAKs) receiver explicitly tells sender that pkt had errorsSender retransmits pkt on receipt of NAK
New mechanisms in rdt20 (beyond rdt10)Error detectionReceiver feedback control msgs (ACKNAK) rcvr-gtsender
Transport Layer 3-26
Rdt20 FSM specification
Wait for call from above
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
sender
receiversnkpkt = make_pkt(data checksum)udt_send(sndpkt)
rdt_send(data)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)Λ
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-7
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-8
MultiplexingDemultiplexing
host 1
= process= socket
1 host 1 or more processes1 process 1 or more socketsTransport layer interacts with socket
bull Demultiplexing at rcv hostbull Multiplexing at send host
application
transport
network
link
physical
P1 application
transport
network
link
physical
application
transport
network
link
physical
P2P3 P4P1
host 1 host 2 host 3
Transport Layer 3-9
How Demultiplexing WorksHost receives IP datagrams
Each datagram has source IP address destination IP addressEach datagram carries 1 transport-layer segmentEach segment has source destination port number (recall well-known port numbers for specific applications)
Host uses IP addresses amp port numbers to direct segment to appropriate socket
source port dest port
32 bits
applicationdata
(message)
other header fields
TCPUDP segment format
Transport Layer 3-10
Connectionless Demultiplexing
Create sockets with port numbers
DatagramSocket mySocket1 = new DatagramSocket(12534)
DatagramSocket mySocket2 = new DatagramSocket(12535)
UDP socket identified by two-tuple
(dest IP address dest port number)
When host receives UDP segment
Checks destination port number in segmentDirects UDP segment to socket with that port number
IP datagrams with different source IP addresses andor source port numbers directed to same socket
Transport Layer 3-11
Connection-Oriented Demux
TCP socket identified by 4-tuple
Source IP addressSource port numberDest IP addressDest port number
Recv host uses all four values to direct segment to appropriate socket
Server host may support many simultaneous TCP sockets
Each socket identified by its own 4-tuple
Web servers have different sockets for each connecting client
Non-persistent HTTP will have different socket for each request
Transport Layer 3-12
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-13
UDP User Datagram Protocol [RFC 768]
ldquoNo frillsrdquo ldquobare bonesrdquoInternet transport protocolldquoBest effortrdquo service UDP segments may be
LostDelivered out of order to app
ConnectionlessNo handshaking between UDP sender receiverEach UDP segment handled independently of others
Why is there a UDPNo connection establishment (which can add delay)Simple no connection state at sender receiverSmall segment headerNo congestion control UDP can blast away as fast as desired
Transport Layer 3-14
UDP moreOften used for streaming multimedia apps
Loss tolerantRate sensitive
Other UDP usesDNSSNMP
Reliable transfer over UDP add reliability at application layer
Application-specific error recovery
source port dest port
32 bits
Applicationdata
(message)
UDP segment format
length checksumLength in
bytes of UDPsegmentincluding
header
Transport Layer 3-15
UDP Checksum
SenderTreat segment contents as sequence of 16-bit integersChecksum addition (1rsquos complement sum) of segment contentsSender puts checksum value into UDP checksum field
ReceiverCompute checksum of received segmentCheck if computed checksum equals checksum field value
NO - error detectedYES - no error detected But maybe errors nonetheless More later
Goal detect ldquoerrorsrdquo (eg flipped bits) in transmitted segment
Transport Layer 3-16
Internet Checksum ExampleNote
When adding numbers a carryout from the most significant bit needs to be added to the result
Example add two 16-bit integers
wraparound
sumchecksum
1 1 1 1 0 0 1 1 0 0 1 1 0 0 1 1 01 1 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 0 1 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 1 0 01 0 1 0 0 0 1 0 0 0 1 0 0 0 0 1 1
Transport Layer 3-17
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-18
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Transport Layer 3-19
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-20
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-21
Reliable Data Transfer Getting Started
sendside
receiveside
rdt_send() called from above (eg by app) Passed data to
deliver to receiver upper layer
udt_send() called by rdtto transfer packet over
unreliable channel to receiver
rdt_rcv() called when packet arrives on rcv-side of channel
deliver_data() called by rdt to deliver data to upper
Transport Layer 3-22
Reliable Data Transfer Getting StartedWersquoll
Incrementally develop sender receiver sides of reliable data transfer protocol (rdt)Consider only unidirectional data transfer
But control info will flow on both directionsUse finite state machines (FSM) to specify sender receiver
state1
state2
event causing state transitionactions taken on state transition
state when in this ldquostaterdquo next state
uniquely determined by next event
eventactions
Transport Layer 3-23
Summary of the Protocols
Rdt10 all packets arrive correctlyRdt20 all packets arrive with possible errors only in data packets and introducing ACK and NAK (no error)Rdt21 corrupted ACKsNAKsRdt22 similar to rdt21 remove NAKsRdt30 Allows packets to be lost and errors
Transport Layer 3-24
Rdt10 Reliable Transfer over a Reliable Channel
Underlying channel perfectly reliableNo bit errorsNo loss of packets
Separate FSMs for sender receiverSender sends data into underlying channelReceiver read data from underlying channel
packet = make_pkt(data)udt_send(packet)
Wait for call from above
rdt_send(data)
sender
extract (packetdata)deliver_data(data)
Wait for call from
below
rdt_rcv(packet)
receiver
Transport Layer 3-25
Rdt20 Channel with Bit Errors
Underlying channel may flip bits in packetChecksum to detect bit errors
The question how to recover from errorsAcknowledgements (ACKs) receiver explicitly tells sender that pkt received OKNegative acknowledgements (NAKs) receiver explicitly tells sender that pkt had errorsSender retransmits pkt on receipt of NAK
New mechanisms in rdt20 (beyond rdt10)Error detectionReceiver feedback control msgs (ACKNAK) rcvr-gtsender
Transport Layer 3-26
Rdt20 FSM specification
Wait for call from above
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
sender
receiversnkpkt = make_pkt(data checksum)udt_send(sndpkt)
rdt_send(data)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)Λ
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-8
MultiplexingDemultiplexing
host 1
= process= socket
1 host 1 or more processes1 process 1 or more socketsTransport layer interacts with socket
bull Demultiplexing at rcv hostbull Multiplexing at send host
application
transport
network
link
physical
P1 application
transport
network
link
physical
application
transport
network
link
physical
P2P3 P4P1
host 1 host 2 host 3
Transport Layer 3-9
How Demultiplexing WorksHost receives IP datagrams
Each datagram has source IP address destination IP addressEach datagram carries 1 transport-layer segmentEach segment has source destination port number (recall well-known port numbers for specific applications)
Host uses IP addresses amp port numbers to direct segment to appropriate socket
source port dest port
32 bits
applicationdata
(message)
other header fields
TCPUDP segment format
Transport Layer 3-10
Connectionless Demultiplexing
Create sockets with port numbers
DatagramSocket mySocket1 = new DatagramSocket(12534)
DatagramSocket mySocket2 = new DatagramSocket(12535)
UDP socket identified by two-tuple
(dest IP address dest port number)
When host receives UDP segment
Checks destination port number in segmentDirects UDP segment to socket with that port number
IP datagrams with different source IP addresses andor source port numbers directed to same socket
Transport Layer 3-11
Connection-Oriented Demux
TCP socket identified by 4-tuple
Source IP addressSource port numberDest IP addressDest port number
Recv host uses all four values to direct segment to appropriate socket
Server host may support many simultaneous TCP sockets
Each socket identified by its own 4-tuple
Web servers have different sockets for each connecting client
Non-persistent HTTP will have different socket for each request
Transport Layer 3-12
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-13
UDP User Datagram Protocol [RFC 768]
ldquoNo frillsrdquo ldquobare bonesrdquoInternet transport protocolldquoBest effortrdquo service UDP segments may be
LostDelivered out of order to app
ConnectionlessNo handshaking between UDP sender receiverEach UDP segment handled independently of others
Why is there a UDPNo connection establishment (which can add delay)Simple no connection state at sender receiverSmall segment headerNo congestion control UDP can blast away as fast as desired
Transport Layer 3-14
UDP moreOften used for streaming multimedia apps
Loss tolerantRate sensitive
Other UDP usesDNSSNMP
Reliable transfer over UDP add reliability at application layer
Application-specific error recovery
source port dest port
32 bits
Applicationdata
(message)
UDP segment format
length checksumLength in
bytes of UDPsegmentincluding
header
Transport Layer 3-15
UDP Checksum
SenderTreat segment contents as sequence of 16-bit integersChecksum addition (1rsquos complement sum) of segment contentsSender puts checksum value into UDP checksum field
ReceiverCompute checksum of received segmentCheck if computed checksum equals checksum field value
NO - error detectedYES - no error detected But maybe errors nonetheless More later
Goal detect ldquoerrorsrdquo (eg flipped bits) in transmitted segment
Transport Layer 3-16
Internet Checksum ExampleNote
When adding numbers a carryout from the most significant bit needs to be added to the result
Example add two 16-bit integers
wraparound
sumchecksum
1 1 1 1 0 0 1 1 0 0 1 1 0 0 1 1 01 1 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 0 1 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 1 0 01 0 1 0 0 0 1 0 0 0 1 0 0 0 0 1 1
Transport Layer 3-17
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-18
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Transport Layer 3-19
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-20
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-21
Reliable Data Transfer Getting Started
sendside
receiveside
rdt_send() called from above (eg by app) Passed data to
deliver to receiver upper layer
udt_send() called by rdtto transfer packet over
unreliable channel to receiver
rdt_rcv() called when packet arrives on rcv-side of channel
deliver_data() called by rdt to deliver data to upper
Transport Layer 3-22
Reliable Data Transfer Getting StartedWersquoll
Incrementally develop sender receiver sides of reliable data transfer protocol (rdt)Consider only unidirectional data transfer
But control info will flow on both directionsUse finite state machines (FSM) to specify sender receiver
state1
state2
event causing state transitionactions taken on state transition
state when in this ldquostaterdquo next state
uniquely determined by next event
eventactions
Transport Layer 3-23
Summary of the Protocols
Rdt10 all packets arrive correctlyRdt20 all packets arrive with possible errors only in data packets and introducing ACK and NAK (no error)Rdt21 corrupted ACKsNAKsRdt22 similar to rdt21 remove NAKsRdt30 Allows packets to be lost and errors
Transport Layer 3-24
Rdt10 Reliable Transfer over a Reliable Channel
Underlying channel perfectly reliableNo bit errorsNo loss of packets
Separate FSMs for sender receiverSender sends data into underlying channelReceiver read data from underlying channel
packet = make_pkt(data)udt_send(packet)
Wait for call from above
rdt_send(data)
sender
extract (packetdata)deliver_data(data)
Wait for call from
below
rdt_rcv(packet)
receiver
Transport Layer 3-25
Rdt20 Channel with Bit Errors
Underlying channel may flip bits in packetChecksum to detect bit errors
The question how to recover from errorsAcknowledgements (ACKs) receiver explicitly tells sender that pkt received OKNegative acknowledgements (NAKs) receiver explicitly tells sender that pkt had errorsSender retransmits pkt on receipt of NAK
New mechanisms in rdt20 (beyond rdt10)Error detectionReceiver feedback control msgs (ACKNAK) rcvr-gtsender
Transport Layer 3-26
Rdt20 FSM specification
Wait for call from above
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
sender
receiversnkpkt = make_pkt(data checksum)udt_send(sndpkt)
rdt_send(data)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)Λ
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-9
How Demultiplexing WorksHost receives IP datagrams
Each datagram has source IP address destination IP addressEach datagram carries 1 transport-layer segmentEach segment has source destination port number (recall well-known port numbers for specific applications)
Host uses IP addresses amp port numbers to direct segment to appropriate socket
source port dest port
32 bits
applicationdata
(message)
other header fields
TCPUDP segment format
Transport Layer 3-10
Connectionless Demultiplexing
Create sockets with port numbers
DatagramSocket mySocket1 = new DatagramSocket(12534)
DatagramSocket mySocket2 = new DatagramSocket(12535)
UDP socket identified by two-tuple
(dest IP address dest port number)
When host receives UDP segment
Checks destination port number in segmentDirects UDP segment to socket with that port number
IP datagrams with different source IP addresses andor source port numbers directed to same socket
Transport Layer 3-11
Connection-Oriented Demux
TCP socket identified by 4-tuple
Source IP addressSource port numberDest IP addressDest port number
Recv host uses all four values to direct segment to appropriate socket
Server host may support many simultaneous TCP sockets
Each socket identified by its own 4-tuple
Web servers have different sockets for each connecting client
Non-persistent HTTP will have different socket for each request
Transport Layer 3-12
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-13
UDP User Datagram Protocol [RFC 768]
ldquoNo frillsrdquo ldquobare bonesrdquoInternet transport protocolldquoBest effortrdquo service UDP segments may be
LostDelivered out of order to app
ConnectionlessNo handshaking between UDP sender receiverEach UDP segment handled independently of others
Why is there a UDPNo connection establishment (which can add delay)Simple no connection state at sender receiverSmall segment headerNo congestion control UDP can blast away as fast as desired
Transport Layer 3-14
UDP moreOften used for streaming multimedia apps
Loss tolerantRate sensitive
Other UDP usesDNSSNMP
Reliable transfer over UDP add reliability at application layer
Application-specific error recovery
source port dest port
32 bits
Applicationdata
(message)
UDP segment format
length checksumLength in
bytes of UDPsegmentincluding
header
Transport Layer 3-15
UDP Checksum
SenderTreat segment contents as sequence of 16-bit integersChecksum addition (1rsquos complement sum) of segment contentsSender puts checksum value into UDP checksum field
ReceiverCompute checksum of received segmentCheck if computed checksum equals checksum field value
NO - error detectedYES - no error detected But maybe errors nonetheless More later
Goal detect ldquoerrorsrdquo (eg flipped bits) in transmitted segment
Transport Layer 3-16
Internet Checksum ExampleNote
When adding numbers a carryout from the most significant bit needs to be added to the result
Example add two 16-bit integers
wraparound
sumchecksum
1 1 1 1 0 0 1 1 0 0 1 1 0 0 1 1 01 1 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 0 1 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 1 0 01 0 1 0 0 0 1 0 0 0 1 0 0 0 0 1 1
Transport Layer 3-17
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-18
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Transport Layer 3-19
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-20
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-21
Reliable Data Transfer Getting Started
sendside
receiveside
rdt_send() called from above (eg by app) Passed data to
deliver to receiver upper layer
udt_send() called by rdtto transfer packet over
unreliable channel to receiver
rdt_rcv() called when packet arrives on rcv-side of channel
deliver_data() called by rdt to deliver data to upper
Transport Layer 3-22
Reliable Data Transfer Getting StartedWersquoll
Incrementally develop sender receiver sides of reliable data transfer protocol (rdt)Consider only unidirectional data transfer
But control info will flow on both directionsUse finite state machines (FSM) to specify sender receiver
state1
state2
event causing state transitionactions taken on state transition
state when in this ldquostaterdquo next state
uniquely determined by next event
eventactions
Transport Layer 3-23
Summary of the Protocols
Rdt10 all packets arrive correctlyRdt20 all packets arrive with possible errors only in data packets and introducing ACK and NAK (no error)Rdt21 corrupted ACKsNAKsRdt22 similar to rdt21 remove NAKsRdt30 Allows packets to be lost and errors
Transport Layer 3-24
Rdt10 Reliable Transfer over a Reliable Channel
Underlying channel perfectly reliableNo bit errorsNo loss of packets
Separate FSMs for sender receiverSender sends data into underlying channelReceiver read data from underlying channel
packet = make_pkt(data)udt_send(packet)
Wait for call from above
rdt_send(data)
sender
extract (packetdata)deliver_data(data)
Wait for call from
below
rdt_rcv(packet)
receiver
Transport Layer 3-25
Rdt20 Channel with Bit Errors
Underlying channel may flip bits in packetChecksum to detect bit errors
The question how to recover from errorsAcknowledgements (ACKs) receiver explicitly tells sender that pkt received OKNegative acknowledgements (NAKs) receiver explicitly tells sender that pkt had errorsSender retransmits pkt on receipt of NAK
New mechanisms in rdt20 (beyond rdt10)Error detectionReceiver feedback control msgs (ACKNAK) rcvr-gtsender
Transport Layer 3-26
Rdt20 FSM specification
Wait for call from above
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
sender
receiversnkpkt = make_pkt(data checksum)udt_send(sndpkt)
rdt_send(data)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)Λ
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-10
Connectionless Demultiplexing
Create sockets with port numbers
DatagramSocket mySocket1 = new DatagramSocket(12534)
DatagramSocket mySocket2 = new DatagramSocket(12535)
UDP socket identified by two-tuple
(dest IP address dest port number)
When host receives UDP segment
Checks destination port number in segmentDirects UDP segment to socket with that port number
IP datagrams with different source IP addresses andor source port numbers directed to same socket
Transport Layer 3-11
Connection-Oriented Demux
TCP socket identified by 4-tuple
Source IP addressSource port numberDest IP addressDest port number
Recv host uses all four values to direct segment to appropriate socket
Server host may support many simultaneous TCP sockets
Each socket identified by its own 4-tuple
Web servers have different sockets for each connecting client
Non-persistent HTTP will have different socket for each request
Transport Layer 3-12
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-13
UDP User Datagram Protocol [RFC 768]
ldquoNo frillsrdquo ldquobare bonesrdquoInternet transport protocolldquoBest effortrdquo service UDP segments may be
LostDelivered out of order to app
ConnectionlessNo handshaking between UDP sender receiverEach UDP segment handled independently of others
Why is there a UDPNo connection establishment (which can add delay)Simple no connection state at sender receiverSmall segment headerNo congestion control UDP can blast away as fast as desired
Transport Layer 3-14
UDP moreOften used for streaming multimedia apps
Loss tolerantRate sensitive
Other UDP usesDNSSNMP
Reliable transfer over UDP add reliability at application layer
Application-specific error recovery
source port dest port
32 bits
Applicationdata
(message)
UDP segment format
length checksumLength in
bytes of UDPsegmentincluding
header
Transport Layer 3-15
UDP Checksum
SenderTreat segment contents as sequence of 16-bit integersChecksum addition (1rsquos complement sum) of segment contentsSender puts checksum value into UDP checksum field
ReceiverCompute checksum of received segmentCheck if computed checksum equals checksum field value
NO - error detectedYES - no error detected But maybe errors nonetheless More later
Goal detect ldquoerrorsrdquo (eg flipped bits) in transmitted segment
Transport Layer 3-16
Internet Checksum ExampleNote
When adding numbers a carryout from the most significant bit needs to be added to the result
Example add two 16-bit integers
wraparound
sumchecksum
1 1 1 1 0 0 1 1 0 0 1 1 0 0 1 1 01 1 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 0 1 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 1 0 01 0 1 0 0 0 1 0 0 0 1 0 0 0 0 1 1
Transport Layer 3-17
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-18
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Transport Layer 3-19
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-20
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-21
Reliable Data Transfer Getting Started
sendside
receiveside
rdt_send() called from above (eg by app) Passed data to
deliver to receiver upper layer
udt_send() called by rdtto transfer packet over
unreliable channel to receiver
rdt_rcv() called when packet arrives on rcv-side of channel
deliver_data() called by rdt to deliver data to upper
Transport Layer 3-22
Reliable Data Transfer Getting StartedWersquoll
Incrementally develop sender receiver sides of reliable data transfer protocol (rdt)Consider only unidirectional data transfer
But control info will flow on both directionsUse finite state machines (FSM) to specify sender receiver
state1
state2
event causing state transitionactions taken on state transition
state when in this ldquostaterdquo next state
uniquely determined by next event
eventactions
Transport Layer 3-23
Summary of the Protocols
Rdt10 all packets arrive correctlyRdt20 all packets arrive with possible errors only in data packets and introducing ACK and NAK (no error)Rdt21 corrupted ACKsNAKsRdt22 similar to rdt21 remove NAKsRdt30 Allows packets to be lost and errors
Transport Layer 3-24
Rdt10 Reliable Transfer over a Reliable Channel
Underlying channel perfectly reliableNo bit errorsNo loss of packets
Separate FSMs for sender receiverSender sends data into underlying channelReceiver read data from underlying channel
packet = make_pkt(data)udt_send(packet)
Wait for call from above
rdt_send(data)
sender
extract (packetdata)deliver_data(data)
Wait for call from
below
rdt_rcv(packet)
receiver
Transport Layer 3-25
Rdt20 Channel with Bit Errors
Underlying channel may flip bits in packetChecksum to detect bit errors
The question how to recover from errorsAcknowledgements (ACKs) receiver explicitly tells sender that pkt received OKNegative acknowledgements (NAKs) receiver explicitly tells sender that pkt had errorsSender retransmits pkt on receipt of NAK
New mechanisms in rdt20 (beyond rdt10)Error detectionReceiver feedback control msgs (ACKNAK) rcvr-gtsender
Transport Layer 3-26
Rdt20 FSM specification
Wait for call from above
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
sender
receiversnkpkt = make_pkt(data checksum)udt_send(sndpkt)
rdt_send(data)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)Λ
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-11
Connection-Oriented Demux
TCP socket identified by 4-tuple
Source IP addressSource port numberDest IP addressDest port number
Recv host uses all four values to direct segment to appropriate socket
Server host may support many simultaneous TCP sockets
Each socket identified by its own 4-tuple
Web servers have different sockets for each connecting client
Non-persistent HTTP will have different socket for each request
Transport Layer 3-12
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-13
UDP User Datagram Protocol [RFC 768]
ldquoNo frillsrdquo ldquobare bonesrdquoInternet transport protocolldquoBest effortrdquo service UDP segments may be
LostDelivered out of order to app
ConnectionlessNo handshaking between UDP sender receiverEach UDP segment handled independently of others
Why is there a UDPNo connection establishment (which can add delay)Simple no connection state at sender receiverSmall segment headerNo congestion control UDP can blast away as fast as desired
Transport Layer 3-14
UDP moreOften used for streaming multimedia apps
Loss tolerantRate sensitive
Other UDP usesDNSSNMP
Reliable transfer over UDP add reliability at application layer
Application-specific error recovery
source port dest port
32 bits
Applicationdata
(message)
UDP segment format
length checksumLength in
bytes of UDPsegmentincluding
header
Transport Layer 3-15
UDP Checksum
SenderTreat segment contents as sequence of 16-bit integersChecksum addition (1rsquos complement sum) of segment contentsSender puts checksum value into UDP checksum field
ReceiverCompute checksum of received segmentCheck if computed checksum equals checksum field value
NO - error detectedYES - no error detected But maybe errors nonetheless More later
Goal detect ldquoerrorsrdquo (eg flipped bits) in transmitted segment
Transport Layer 3-16
Internet Checksum ExampleNote
When adding numbers a carryout from the most significant bit needs to be added to the result
Example add two 16-bit integers
wraparound
sumchecksum
1 1 1 1 0 0 1 1 0 0 1 1 0 0 1 1 01 1 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 0 1 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 1 0 01 0 1 0 0 0 1 0 0 0 1 0 0 0 0 1 1
Transport Layer 3-17
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-18
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Transport Layer 3-19
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-20
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-21
Reliable Data Transfer Getting Started
sendside
receiveside
rdt_send() called from above (eg by app) Passed data to
deliver to receiver upper layer
udt_send() called by rdtto transfer packet over
unreliable channel to receiver
rdt_rcv() called when packet arrives on rcv-side of channel
deliver_data() called by rdt to deliver data to upper
Transport Layer 3-22
Reliable Data Transfer Getting StartedWersquoll
Incrementally develop sender receiver sides of reliable data transfer protocol (rdt)Consider only unidirectional data transfer
But control info will flow on both directionsUse finite state machines (FSM) to specify sender receiver
state1
state2
event causing state transitionactions taken on state transition
state when in this ldquostaterdquo next state
uniquely determined by next event
eventactions
Transport Layer 3-23
Summary of the Protocols
Rdt10 all packets arrive correctlyRdt20 all packets arrive with possible errors only in data packets and introducing ACK and NAK (no error)Rdt21 corrupted ACKsNAKsRdt22 similar to rdt21 remove NAKsRdt30 Allows packets to be lost and errors
Transport Layer 3-24
Rdt10 Reliable Transfer over a Reliable Channel
Underlying channel perfectly reliableNo bit errorsNo loss of packets
Separate FSMs for sender receiverSender sends data into underlying channelReceiver read data from underlying channel
packet = make_pkt(data)udt_send(packet)
Wait for call from above
rdt_send(data)
sender
extract (packetdata)deliver_data(data)
Wait for call from
below
rdt_rcv(packet)
receiver
Transport Layer 3-25
Rdt20 Channel with Bit Errors
Underlying channel may flip bits in packetChecksum to detect bit errors
The question how to recover from errorsAcknowledgements (ACKs) receiver explicitly tells sender that pkt received OKNegative acknowledgements (NAKs) receiver explicitly tells sender that pkt had errorsSender retransmits pkt on receipt of NAK
New mechanisms in rdt20 (beyond rdt10)Error detectionReceiver feedback control msgs (ACKNAK) rcvr-gtsender
Transport Layer 3-26
Rdt20 FSM specification
Wait for call from above
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
sender
receiversnkpkt = make_pkt(data checksum)udt_send(sndpkt)
rdt_send(data)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)Λ
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-12
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-13
UDP User Datagram Protocol [RFC 768]
ldquoNo frillsrdquo ldquobare bonesrdquoInternet transport protocolldquoBest effortrdquo service UDP segments may be
LostDelivered out of order to app
ConnectionlessNo handshaking between UDP sender receiverEach UDP segment handled independently of others
Why is there a UDPNo connection establishment (which can add delay)Simple no connection state at sender receiverSmall segment headerNo congestion control UDP can blast away as fast as desired
Transport Layer 3-14
UDP moreOften used for streaming multimedia apps
Loss tolerantRate sensitive
Other UDP usesDNSSNMP
Reliable transfer over UDP add reliability at application layer
Application-specific error recovery
source port dest port
32 bits
Applicationdata
(message)
UDP segment format
length checksumLength in
bytes of UDPsegmentincluding
header
Transport Layer 3-15
UDP Checksum
SenderTreat segment contents as sequence of 16-bit integersChecksum addition (1rsquos complement sum) of segment contentsSender puts checksum value into UDP checksum field
ReceiverCompute checksum of received segmentCheck if computed checksum equals checksum field value
NO - error detectedYES - no error detected But maybe errors nonetheless More later
Goal detect ldquoerrorsrdquo (eg flipped bits) in transmitted segment
Transport Layer 3-16
Internet Checksum ExampleNote
When adding numbers a carryout from the most significant bit needs to be added to the result
Example add two 16-bit integers
wraparound
sumchecksum
1 1 1 1 0 0 1 1 0 0 1 1 0 0 1 1 01 1 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 0 1 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 1 0 01 0 1 0 0 0 1 0 0 0 1 0 0 0 0 1 1
Transport Layer 3-17
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-18
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Transport Layer 3-19
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-20
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-21
Reliable Data Transfer Getting Started
sendside
receiveside
rdt_send() called from above (eg by app) Passed data to
deliver to receiver upper layer
udt_send() called by rdtto transfer packet over
unreliable channel to receiver
rdt_rcv() called when packet arrives on rcv-side of channel
deliver_data() called by rdt to deliver data to upper
Transport Layer 3-22
Reliable Data Transfer Getting StartedWersquoll
Incrementally develop sender receiver sides of reliable data transfer protocol (rdt)Consider only unidirectional data transfer
But control info will flow on both directionsUse finite state machines (FSM) to specify sender receiver
state1
state2
event causing state transitionactions taken on state transition
state when in this ldquostaterdquo next state
uniquely determined by next event
eventactions
Transport Layer 3-23
Summary of the Protocols
Rdt10 all packets arrive correctlyRdt20 all packets arrive with possible errors only in data packets and introducing ACK and NAK (no error)Rdt21 corrupted ACKsNAKsRdt22 similar to rdt21 remove NAKsRdt30 Allows packets to be lost and errors
Transport Layer 3-24
Rdt10 Reliable Transfer over a Reliable Channel
Underlying channel perfectly reliableNo bit errorsNo loss of packets
Separate FSMs for sender receiverSender sends data into underlying channelReceiver read data from underlying channel
packet = make_pkt(data)udt_send(packet)
Wait for call from above
rdt_send(data)
sender
extract (packetdata)deliver_data(data)
Wait for call from
below
rdt_rcv(packet)
receiver
Transport Layer 3-25
Rdt20 Channel with Bit Errors
Underlying channel may flip bits in packetChecksum to detect bit errors
The question how to recover from errorsAcknowledgements (ACKs) receiver explicitly tells sender that pkt received OKNegative acknowledgements (NAKs) receiver explicitly tells sender that pkt had errorsSender retransmits pkt on receipt of NAK
New mechanisms in rdt20 (beyond rdt10)Error detectionReceiver feedback control msgs (ACKNAK) rcvr-gtsender
Transport Layer 3-26
Rdt20 FSM specification
Wait for call from above
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
sender
receiversnkpkt = make_pkt(data checksum)udt_send(sndpkt)
rdt_send(data)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)Λ
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-13
UDP User Datagram Protocol [RFC 768]
ldquoNo frillsrdquo ldquobare bonesrdquoInternet transport protocolldquoBest effortrdquo service UDP segments may be
LostDelivered out of order to app
ConnectionlessNo handshaking between UDP sender receiverEach UDP segment handled independently of others
Why is there a UDPNo connection establishment (which can add delay)Simple no connection state at sender receiverSmall segment headerNo congestion control UDP can blast away as fast as desired
Transport Layer 3-14
UDP moreOften used for streaming multimedia apps
Loss tolerantRate sensitive
Other UDP usesDNSSNMP
Reliable transfer over UDP add reliability at application layer
Application-specific error recovery
source port dest port
32 bits
Applicationdata
(message)
UDP segment format
length checksumLength in
bytes of UDPsegmentincluding
header
Transport Layer 3-15
UDP Checksum
SenderTreat segment contents as sequence of 16-bit integersChecksum addition (1rsquos complement sum) of segment contentsSender puts checksum value into UDP checksum field
ReceiverCompute checksum of received segmentCheck if computed checksum equals checksum field value
NO - error detectedYES - no error detected But maybe errors nonetheless More later
Goal detect ldquoerrorsrdquo (eg flipped bits) in transmitted segment
Transport Layer 3-16
Internet Checksum ExampleNote
When adding numbers a carryout from the most significant bit needs to be added to the result
Example add two 16-bit integers
wraparound
sumchecksum
1 1 1 1 0 0 1 1 0 0 1 1 0 0 1 1 01 1 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 0 1 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 1 0 01 0 1 0 0 0 1 0 0 0 1 0 0 0 0 1 1
Transport Layer 3-17
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-18
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Transport Layer 3-19
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-20
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-21
Reliable Data Transfer Getting Started
sendside
receiveside
rdt_send() called from above (eg by app) Passed data to
deliver to receiver upper layer
udt_send() called by rdtto transfer packet over
unreliable channel to receiver
rdt_rcv() called when packet arrives on rcv-side of channel
deliver_data() called by rdt to deliver data to upper
Transport Layer 3-22
Reliable Data Transfer Getting StartedWersquoll
Incrementally develop sender receiver sides of reliable data transfer protocol (rdt)Consider only unidirectional data transfer
But control info will flow on both directionsUse finite state machines (FSM) to specify sender receiver
state1
state2
event causing state transitionactions taken on state transition
state when in this ldquostaterdquo next state
uniquely determined by next event
eventactions
Transport Layer 3-23
Summary of the Protocols
Rdt10 all packets arrive correctlyRdt20 all packets arrive with possible errors only in data packets and introducing ACK and NAK (no error)Rdt21 corrupted ACKsNAKsRdt22 similar to rdt21 remove NAKsRdt30 Allows packets to be lost and errors
Transport Layer 3-24
Rdt10 Reliable Transfer over a Reliable Channel
Underlying channel perfectly reliableNo bit errorsNo loss of packets
Separate FSMs for sender receiverSender sends data into underlying channelReceiver read data from underlying channel
packet = make_pkt(data)udt_send(packet)
Wait for call from above
rdt_send(data)
sender
extract (packetdata)deliver_data(data)
Wait for call from
below
rdt_rcv(packet)
receiver
Transport Layer 3-25
Rdt20 Channel with Bit Errors
Underlying channel may flip bits in packetChecksum to detect bit errors
The question how to recover from errorsAcknowledgements (ACKs) receiver explicitly tells sender that pkt received OKNegative acknowledgements (NAKs) receiver explicitly tells sender that pkt had errorsSender retransmits pkt on receipt of NAK
New mechanisms in rdt20 (beyond rdt10)Error detectionReceiver feedback control msgs (ACKNAK) rcvr-gtsender
Transport Layer 3-26
Rdt20 FSM specification
Wait for call from above
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
sender
receiversnkpkt = make_pkt(data checksum)udt_send(sndpkt)
rdt_send(data)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)Λ
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-14
UDP moreOften used for streaming multimedia apps
Loss tolerantRate sensitive
Other UDP usesDNSSNMP
Reliable transfer over UDP add reliability at application layer
Application-specific error recovery
source port dest port
32 bits
Applicationdata
(message)
UDP segment format
length checksumLength in
bytes of UDPsegmentincluding
header
Transport Layer 3-15
UDP Checksum
SenderTreat segment contents as sequence of 16-bit integersChecksum addition (1rsquos complement sum) of segment contentsSender puts checksum value into UDP checksum field
ReceiverCompute checksum of received segmentCheck if computed checksum equals checksum field value
NO - error detectedYES - no error detected But maybe errors nonetheless More later
Goal detect ldquoerrorsrdquo (eg flipped bits) in transmitted segment
Transport Layer 3-16
Internet Checksum ExampleNote
When adding numbers a carryout from the most significant bit needs to be added to the result
Example add two 16-bit integers
wraparound
sumchecksum
1 1 1 1 0 0 1 1 0 0 1 1 0 0 1 1 01 1 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 0 1 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 1 0 01 0 1 0 0 0 1 0 0 0 1 0 0 0 0 1 1
Transport Layer 3-17
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-18
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Transport Layer 3-19
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-20
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-21
Reliable Data Transfer Getting Started
sendside
receiveside
rdt_send() called from above (eg by app) Passed data to
deliver to receiver upper layer
udt_send() called by rdtto transfer packet over
unreliable channel to receiver
rdt_rcv() called when packet arrives on rcv-side of channel
deliver_data() called by rdt to deliver data to upper
Transport Layer 3-22
Reliable Data Transfer Getting StartedWersquoll
Incrementally develop sender receiver sides of reliable data transfer protocol (rdt)Consider only unidirectional data transfer
But control info will flow on both directionsUse finite state machines (FSM) to specify sender receiver
state1
state2
event causing state transitionactions taken on state transition
state when in this ldquostaterdquo next state
uniquely determined by next event
eventactions
Transport Layer 3-23
Summary of the Protocols
Rdt10 all packets arrive correctlyRdt20 all packets arrive with possible errors only in data packets and introducing ACK and NAK (no error)Rdt21 corrupted ACKsNAKsRdt22 similar to rdt21 remove NAKsRdt30 Allows packets to be lost and errors
Transport Layer 3-24
Rdt10 Reliable Transfer over a Reliable Channel
Underlying channel perfectly reliableNo bit errorsNo loss of packets
Separate FSMs for sender receiverSender sends data into underlying channelReceiver read data from underlying channel
packet = make_pkt(data)udt_send(packet)
Wait for call from above
rdt_send(data)
sender
extract (packetdata)deliver_data(data)
Wait for call from
below
rdt_rcv(packet)
receiver
Transport Layer 3-25
Rdt20 Channel with Bit Errors
Underlying channel may flip bits in packetChecksum to detect bit errors
The question how to recover from errorsAcknowledgements (ACKs) receiver explicitly tells sender that pkt received OKNegative acknowledgements (NAKs) receiver explicitly tells sender that pkt had errorsSender retransmits pkt on receipt of NAK
New mechanisms in rdt20 (beyond rdt10)Error detectionReceiver feedback control msgs (ACKNAK) rcvr-gtsender
Transport Layer 3-26
Rdt20 FSM specification
Wait for call from above
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
sender
receiversnkpkt = make_pkt(data checksum)udt_send(sndpkt)
rdt_send(data)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)Λ
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-15
UDP Checksum
SenderTreat segment contents as sequence of 16-bit integersChecksum addition (1rsquos complement sum) of segment contentsSender puts checksum value into UDP checksum field
ReceiverCompute checksum of received segmentCheck if computed checksum equals checksum field value
NO - error detectedYES - no error detected But maybe errors nonetheless More later
Goal detect ldquoerrorsrdquo (eg flipped bits) in transmitted segment
Transport Layer 3-16
Internet Checksum ExampleNote
When adding numbers a carryout from the most significant bit needs to be added to the result
Example add two 16-bit integers
wraparound
sumchecksum
1 1 1 1 0 0 1 1 0 0 1 1 0 0 1 1 01 1 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 0 1 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 1 0 01 0 1 0 0 0 1 0 0 0 1 0 0 0 0 1 1
Transport Layer 3-17
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-18
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Transport Layer 3-19
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-20
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-21
Reliable Data Transfer Getting Started
sendside
receiveside
rdt_send() called from above (eg by app) Passed data to
deliver to receiver upper layer
udt_send() called by rdtto transfer packet over
unreliable channel to receiver
rdt_rcv() called when packet arrives on rcv-side of channel
deliver_data() called by rdt to deliver data to upper
Transport Layer 3-22
Reliable Data Transfer Getting StartedWersquoll
Incrementally develop sender receiver sides of reliable data transfer protocol (rdt)Consider only unidirectional data transfer
But control info will flow on both directionsUse finite state machines (FSM) to specify sender receiver
state1
state2
event causing state transitionactions taken on state transition
state when in this ldquostaterdquo next state
uniquely determined by next event
eventactions
Transport Layer 3-23
Summary of the Protocols
Rdt10 all packets arrive correctlyRdt20 all packets arrive with possible errors only in data packets and introducing ACK and NAK (no error)Rdt21 corrupted ACKsNAKsRdt22 similar to rdt21 remove NAKsRdt30 Allows packets to be lost and errors
Transport Layer 3-24
Rdt10 Reliable Transfer over a Reliable Channel
Underlying channel perfectly reliableNo bit errorsNo loss of packets
Separate FSMs for sender receiverSender sends data into underlying channelReceiver read data from underlying channel
packet = make_pkt(data)udt_send(packet)
Wait for call from above
rdt_send(data)
sender
extract (packetdata)deliver_data(data)
Wait for call from
below
rdt_rcv(packet)
receiver
Transport Layer 3-25
Rdt20 Channel with Bit Errors
Underlying channel may flip bits in packetChecksum to detect bit errors
The question how to recover from errorsAcknowledgements (ACKs) receiver explicitly tells sender that pkt received OKNegative acknowledgements (NAKs) receiver explicitly tells sender that pkt had errorsSender retransmits pkt on receipt of NAK
New mechanisms in rdt20 (beyond rdt10)Error detectionReceiver feedback control msgs (ACKNAK) rcvr-gtsender
Transport Layer 3-26
Rdt20 FSM specification
Wait for call from above
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
sender
receiversnkpkt = make_pkt(data checksum)udt_send(sndpkt)
rdt_send(data)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)Λ
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-16
Internet Checksum ExampleNote
When adding numbers a carryout from the most significant bit needs to be added to the result
Example add two 16-bit integers
wraparound
sumchecksum
1 1 1 1 0 0 1 1 0 0 1 1 0 0 1 1 01 1 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 0 1 1
1 1 0 1 1 1 0 1 1 1 0 1 1 1 1 0 01 0 1 0 0 0 1 0 0 0 1 0 0 0 0 1 1
Transport Layer 3-17
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-18
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Transport Layer 3-19
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-20
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-21
Reliable Data Transfer Getting Started
sendside
receiveside
rdt_send() called from above (eg by app) Passed data to
deliver to receiver upper layer
udt_send() called by rdtto transfer packet over
unreliable channel to receiver
rdt_rcv() called when packet arrives on rcv-side of channel
deliver_data() called by rdt to deliver data to upper
Transport Layer 3-22
Reliable Data Transfer Getting StartedWersquoll
Incrementally develop sender receiver sides of reliable data transfer protocol (rdt)Consider only unidirectional data transfer
But control info will flow on both directionsUse finite state machines (FSM) to specify sender receiver
state1
state2
event causing state transitionactions taken on state transition
state when in this ldquostaterdquo next state
uniquely determined by next event
eventactions
Transport Layer 3-23
Summary of the Protocols
Rdt10 all packets arrive correctlyRdt20 all packets arrive with possible errors only in data packets and introducing ACK and NAK (no error)Rdt21 corrupted ACKsNAKsRdt22 similar to rdt21 remove NAKsRdt30 Allows packets to be lost and errors
Transport Layer 3-24
Rdt10 Reliable Transfer over a Reliable Channel
Underlying channel perfectly reliableNo bit errorsNo loss of packets
Separate FSMs for sender receiverSender sends data into underlying channelReceiver read data from underlying channel
packet = make_pkt(data)udt_send(packet)
Wait for call from above
rdt_send(data)
sender
extract (packetdata)deliver_data(data)
Wait for call from
below
rdt_rcv(packet)
receiver
Transport Layer 3-25
Rdt20 Channel with Bit Errors
Underlying channel may flip bits in packetChecksum to detect bit errors
The question how to recover from errorsAcknowledgements (ACKs) receiver explicitly tells sender that pkt received OKNegative acknowledgements (NAKs) receiver explicitly tells sender that pkt had errorsSender retransmits pkt on receipt of NAK
New mechanisms in rdt20 (beyond rdt10)Error detectionReceiver feedback control msgs (ACKNAK) rcvr-gtsender
Transport Layer 3-26
Rdt20 FSM specification
Wait for call from above
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
sender
receiversnkpkt = make_pkt(data checksum)udt_send(sndpkt)
rdt_send(data)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)Λ
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-17
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-18
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Transport Layer 3-19
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-20
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-21
Reliable Data Transfer Getting Started
sendside
receiveside
rdt_send() called from above (eg by app) Passed data to
deliver to receiver upper layer
udt_send() called by rdtto transfer packet over
unreliable channel to receiver
rdt_rcv() called when packet arrives on rcv-side of channel
deliver_data() called by rdt to deliver data to upper
Transport Layer 3-22
Reliable Data Transfer Getting StartedWersquoll
Incrementally develop sender receiver sides of reliable data transfer protocol (rdt)Consider only unidirectional data transfer
But control info will flow on both directionsUse finite state machines (FSM) to specify sender receiver
state1
state2
event causing state transitionactions taken on state transition
state when in this ldquostaterdquo next state
uniquely determined by next event
eventactions
Transport Layer 3-23
Summary of the Protocols
Rdt10 all packets arrive correctlyRdt20 all packets arrive with possible errors only in data packets and introducing ACK and NAK (no error)Rdt21 corrupted ACKsNAKsRdt22 similar to rdt21 remove NAKsRdt30 Allows packets to be lost and errors
Transport Layer 3-24
Rdt10 Reliable Transfer over a Reliable Channel
Underlying channel perfectly reliableNo bit errorsNo loss of packets
Separate FSMs for sender receiverSender sends data into underlying channelReceiver read data from underlying channel
packet = make_pkt(data)udt_send(packet)
Wait for call from above
rdt_send(data)
sender
extract (packetdata)deliver_data(data)
Wait for call from
below
rdt_rcv(packet)
receiver
Transport Layer 3-25
Rdt20 Channel with Bit Errors
Underlying channel may flip bits in packetChecksum to detect bit errors
The question how to recover from errorsAcknowledgements (ACKs) receiver explicitly tells sender that pkt received OKNegative acknowledgements (NAKs) receiver explicitly tells sender that pkt had errorsSender retransmits pkt on receipt of NAK
New mechanisms in rdt20 (beyond rdt10)Error detectionReceiver feedback control msgs (ACKNAK) rcvr-gtsender
Transport Layer 3-26
Rdt20 FSM specification
Wait for call from above
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
sender
receiversnkpkt = make_pkt(data checksum)udt_send(sndpkt)
rdt_send(data)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)Λ
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-18
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Transport Layer 3-19
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-20
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-21
Reliable Data Transfer Getting Started
sendside
receiveside
rdt_send() called from above (eg by app) Passed data to
deliver to receiver upper layer
udt_send() called by rdtto transfer packet over
unreliable channel to receiver
rdt_rcv() called when packet arrives on rcv-side of channel
deliver_data() called by rdt to deliver data to upper
Transport Layer 3-22
Reliable Data Transfer Getting StartedWersquoll
Incrementally develop sender receiver sides of reliable data transfer protocol (rdt)Consider only unidirectional data transfer
But control info will flow on both directionsUse finite state machines (FSM) to specify sender receiver
state1
state2
event causing state transitionactions taken on state transition
state when in this ldquostaterdquo next state
uniquely determined by next event
eventactions
Transport Layer 3-23
Summary of the Protocols
Rdt10 all packets arrive correctlyRdt20 all packets arrive with possible errors only in data packets and introducing ACK and NAK (no error)Rdt21 corrupted ACKsNAKsRdt22 similar to rdt21 remove NAKsRdt30 Allows packets to be lost and errors
Transport Layer 3-24
Rdt10 Reliable Transfer over a Reliable Channel
Underlying channel perfectly reliableNo bit errorsNo loss of packets
Separate FSMs for sender receiverSender sends data into underlying channelReceiver read data from underlying channel
packet = make_pkt(data)udt_send(packet)
Wait for call from above
rdt_send(data)
sender
extract (packetdata)deliver_data(data)
Wait for call from
below
rdt_rcv(packet)
receiver
Transport Layer 3-25
Rdt20 Channel with Bit Errors
Underlying channel may flip bits in packetChecksum to detect bit errors
The question how to recover from errorsAcknowledgements (ACKs) receiver explicitly tells sender that pkt received OKNegative acknowledgements (NAKs) receiver explicitly tells sender that pkt had errorsSender retransmits pkt on receipt of NAK
New mechanisms in rdt20 (beyond rdt10)Error detectionReceiver feedback control msgs (ACKNAK) rcvr-gtsender
Transport Layer 3-26
Rdt20 FSM specification
Wait for call from above
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
sender
receiversnkpkt = make_pkt(data checksum)udt_send(sndpkt)
rdt_send(data)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)Λ
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-19
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-20
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-21
Reliable Data Transfer Getting Started
sendside
receiveside
rdt_send() called from above (eg by app) Passed data to
deliver to receiver upper layer
udt_send() called by rdtto transfer packet over
unreliable channel to receiver
rdt_rcv() called when packet arrives on rcv-side of channel
deliver_data() called by rdt to deliver data to upper
Transport Layer 3-22
Reliable Data Transfer Getting StartedWersquoll
Incrementally develop sender receiver sides of reliable data transfer protocol (rdt)Consider only unidirectional data transfer
But control info will flow on both directionsUse finite state machines (FSM) to specify sender receiver
state1
state2
event causing state transitionactions taken on state transition
state when in this ldquostaterdquo next state
uniquely determined by next event
eventactions
Transport Layer 3-23
Summary of the Protocols
Rdt10 all packets arrive correctlyRdt20 all packets arrive with possible errors only in data packets and introducing ACK and NAK (no error)Rdt21 corrupted ACKsNAKsRdt22 similar to rdt21 remove NAKsRdt30 Allows packets to be lost and errors
Transport Layer 3-24
Rdt10 Reliable Transfer over a Reliable Channel
Underlying channel perfectly reliableNo bit errorsNo loss of packets
Separate FSMs for sender receiverSender sends data into underlying channelReceiver read data from underlying channel
packet = make_pkt(data)udt_send(packet)
Wait for call from above
rdt_send(data)
sender
extract (packetdata)deliver_data(data)
Wait for call from
below
rdt_rcv(packet)
receiver
Transport Layer 3-25
Rdt20 Channel with Bit Errors
Underlying channel may flip bits in packetChecksum to detect bit errors
The question how to recover from errorsAcknowledgements (ACKs) receiver explicitly tells sender that pkt received OKNegative acknowledgements (NAKs) receiver explicitly tells sender that pkt had errorsSender retransmits pkt on receipt of NAK
New mechanisms in rdt20 (beyond rdt10)Error detectionReceiver feedback control msgs (ACKNAK) rcvr-gtsender
Transport Layer 3-26
Rdt20 FSM specification
Wait for call from above
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
sender
receiversnkpkt = make_pkt(data checksum)udt_send(sndpkt)
rdt_send(data)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)Λ
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-20
Principles of Reliable Data TransferImportant in app transport link layersTop-10 list of important networking topics
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol (rdt)
Transport Layer 3-21
Reliable Data Transfer Getting Started
sendside
receiveside
rdt_send() called from above (eg by app) Passed data to
deliver to receiver upper layer
udt_send() called by rdtto transfer packet over
unreliable channel to receiver
rdt_rcv() called when packet arrives on rcv-side of channel
deliver_data() called by rdt to deliver data to upper
Transport Layer 3-22
Reliable Data Transfer Getting StartedWersquoll
Incrementally develop sender receiver sides of reliable data transfer protocol (rdt)Consider only unidirectional data transfer
But control info will flow on both directionsUse finite state machines (FSM) to specify sender receiver
state1
state2
event causing state transitionactions taken on state transition
state when in this ldquostaterdquo next state
uniquely determined by next event
eventactions
Transport Layer 3-23
Summary of the Protocols
Rdt10 all packets arrive correctlyRdt20 all packets arrive with possible errors only in data packets and introducing ACK and NAK (no error)Rdt21 corrupted ACKsNAKsRdt22 similar to rdt21 remove NAKsRdt30 Allows packets to be lost and errors
Transport Layer 3-24
Rdt10 Reliable Transfer over a Reliable Channel
Underlying channel perfectly reliableNo bit errorsNo loss of packets
Separate FSMs for sender receiverSender sends data into underlying channelReceiver read data from underlying channel
packet = make_pkt(data)udt_send(packet)
Wait for call from above
rdt_send(data)
sender
extract (packetdata)deliver_data(data)
Wait for call from
below
rdt_rcv(packet)
receiver
Transport Layer 3-25
Rdt20 Channel with Bit Errors
Underlying channel may flip bits in packetChecksum to detect bit errors
The question how to recover from errorsAcknowledgements (ACKs) receiver explicitly tells sender that pkt received OKNegative acknowledgements (NAKs) receiver explicitly tells sender that pkt had errorsSender retransmits pkt on receipt of NAK
New mechanisms in rdt20 (beyond rdt10)Error detectionReceiver feedback control msgs (ACKNAK) rcvr-gtsender
Transport Layer 3-26
Rdt20 FSM specification
Wait for call from above
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
sender
receiversnkpkt = make_pkt(data checksum)udt_send(sndpkt)
rdt_send(data)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)Λ
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-21
Reliable Data Transfer Getting Started
sendside
receiveside
rdt_send() called from above (eg by app) Passed data to
deliver to receiver upper layer
udt_send() called by rdtto transfer packet over
unreliable channel to receiver
rdt_rcv() called when packet arrives on rcv-side of channel
deliver_data() called by rdt to deliver data to upper
Transport Layer 3-22
Reliable Data Transfer Getting StartedWersquoll
Incrementally develop sender receiver sides of reliable data transfer protocol (rdt)Consider only unidirectional data transfer
But control info will flow on both directionsUse finite state machines (FSM) to specify sender receiver
state1
state2
event causing state transitionactions taken on state transition
state when in this ldquostaterdquo next state
uniquely determined by next event
eventactions
Transport Layer 3-23
Summary of the Protocols
Rdt10 all packets arrive correctlyRdt20 all packets arrive with possible errors only in data packets and introducing ACK and NAK (no error)Rdt21 corrupted ACKsNAKsRdt22 similar to rdt21 remove NAKsRdt30 Allows packets to be lost and errors
Transport Layer 3-24
Rdt10 Reliable Transfer over a Reliable Channel
Underlying channel perfectly reliableNo bit errorsNo loss of packets
Separate FSMs for sender receiverSender sends data into underlying channelReceiver read data from underlying channel
packet = make_pkt(data)udt_send(packet)
Wait for call from above
rdt_send(data)
sender
extract (packetdata)deliver_data(data)
Wait for call from
below
rdt_rcv(packet)
receiver
Transport Layer 3-25
Rdt20 Channel with Bit Errors
Underlying channel may flip bits in packetChecksum to detect bit errors
The question how to recover from errorsAcknowledgements (ACKs) receiver explicitly tells sender that pkt received OKNegative acknowledgements (NAKs) receiver explicitly tells sender that pkt had errorsSender retransmits pkt on receipt of NAK
New mechanisms in rdt20 (beyond rdt10)Error detectionReceiver feedback control msgs (ACKNAK) rcvr-gtsender
Transport Layer 3-26
Rdt20 FSM specification
Wait for call from above
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
sender
receiversnkpkt = make_pkt(data checksum)udt_send(sndpkt)
rdt_send(data)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)Λ
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-22
Reliable Data Transfer Getting StartedWersquoll
Incrementally develop sender receiver sides of reliable data transfer protocol (rdt)Consider only unidirectional data transfer
But control info will flow on both directionsUse finite state machines (FSM) to specify sender receiver
state1
state2
event causing state transitionactions taken on state transition
state when in this ldquostaterdquo next state
uniquely determined by next event
eventactions
Transport Layer 3-23
Summary of the Protocols
Rdt10 all packets arrive correctlyRdt20 all packets arrive with possible errors only in data packets and introducing ACK and NAK (no error)Rdt21 corrupted ACKsNAKsRdt22 similar to rdt21 remove NAKsRdt30 Allows packets to be lost and errors
Transport Layer 3-24
Rdt10 Reliable Transfer over a Reliable Channel
Underlying channel perfectly reliableNo bit errorsNo loss of packets
Separate FSMs for sender receiverSender sends data into underlying channelReceiver read data from underlying channel
packet = make_pkt(data)udt_send(packet)
Wait for call from above
rdt_send(data)
sender
extract (packetdata)deliver_data(data)
Wait for call from
below
rdt_rcv(packet)
receiver
Transport Layer 3-25
Rdt20 Channel with Bit Errors
Underlying channel may flip bits in packetChecksum to detect bit errors
The question how to recover from errorsAcknowledgements (ACKs) receiver explicitly tells sender that pkt received OKNegative acknowledgements (NAKs) receiver explicitly tells sender that pkt had errorsSender retransmits pkt on receipt of NAK
New mechanisms in rdt20 (beyond rdt10)Error detectionReceiver feedback control msgs (ACKNAK) rcvr-gtsender
Transport Layer 3-26
Rdt20 FSM specification
Wait for call from above
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
sender
receiversnkpkt = make_pkt(data checksum)udt_send(sndpkt)
rdt_send(data)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)Λ
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-23
Summary of the Protocols
Rdt10 all packets arrive correctlyRdt20 all packets arrive with possible errors only in data packets and introducing ACK and NAK (no error)Rdt21 corrupted ACKsNAKsRdt22 similar to rdt21 remove NAKsRdt30 Allows packets to be lost and errors
Transport Layer 3-24
Rdt10 Reliable Transfer over a Reliable Channel
Underlying channel perfectly reliableNo bit errorsNo loss of packets
Separate FSMs for sender receiverSender sends data into underlying channelReceiver read data from underlying channel
packet = make_pkt(data)udt_send(packet)
Wait for call from above
rdt_send(data)
sender
extract (packetdata)deliver_data(data)
Wait for call from
below
rdt_rcv(packet)
receiver
Transport Layer 3-25
Rdt20 Channel with Bit Errors
Underlying channel may flip bits in packetChecksum to detect bit errors
The question how to recover from errorsAcknowledgements (ACKs) receiver explicitly tells sender that pkt received OKNegative acknowledgements (NAKs) receiver explicitly tells sender that pkt had errorsSender retransmits pkt on receipt of NAK
New mechanisms in rdt20 (beyond rdt10)Error detectionReceiver feedback control msgs (ACKNAK) rcvr-gtsender
Transport Layer 3-26
Rdt20 FSM specification
Wait for call from above
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
sender
receiversnkpkt = make_pkt(data checksum)udt_send(sndpkt)
rdt_send(data)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)Λ
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-24
Rdt10 Reliable Transfer over a Reliable Channel
Underlying channel perfectly reliableNo bit errorsNo loss of packets
Separate FSMs for sender receiverSender sends data into underlying channelReceiver read data from underlying channel
packet = make_pkt(data)udt_send(packet)
Wait for call from above
rdt_send(data)
sender
extract (packetdata)deliver_data(data)
Wait for call from
below
rdt_rcv(packet)
receiver
Transport Layer 3-25
Rdt20 Channel with Bit Errors
Underlying channel may flip bits in packetChecksum to detect bit errors
The question how to recover from errorsAcknowledgements (ACKs) receiver explicitly tells sender that pkt received OKNegative acknowledgements (NAKs) receiver explicitly tells sender that pkt had errorsSender retransmits pkt on receipt of NAK
New mechanisms in rdt20 (beyond rdt10)Error detectionReceiver feedback control msgs (ACKNAK) rcvr-gtsender
Transport Layer 3-26
Rdt20 FSM specification
Wait for call from above
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
sender
receiversnkpkt = make_pkt(data checksum)udt_send(sndpkt)
rdt_send(data)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)Λ
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-25
Rdt20 Channel with Bit Errors
Underlying channel may flip bits in packetChecksum to detect bit errors
The question how to recover from errorsAcknowledgements (ACKs) receiver explicitly tells sender that pkt received OKNegative acknowledgements (NAKs) receiver explicitly tells sender that pkt had errorsSender retransmits pkt on receipt of NAK
New mechanisms in rdt20 (beyond rdt10)Error detectionReceiver feedback control msgs (ACKNAK) rcvr-gtsender
Transport Layer 3-26
Rdt20 FSM specification
Wait for call from above
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
sender
receiversnkpkt = make_pkt(data checksum)udt_send(sndpkt)
rdt_send(data)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)Λ
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-26
Rdt20 FSM specification
Wait for call from above
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
sender
receiversnkpkt = make_pkt(data checksum)udt_send(sndpkt)
rdt_send(data)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)Λ
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-27
Rdt20 Operation with No Errors
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
sender
receiver
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-28
Rdt20 Error Scenario
Wait for call from above
snkpkt = make_pkt(data checksum)udt_send(sndpkt)
extract(rcvpktdata)deliver_data(data)udt_send(ACK)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp isACK(rcvpkt)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampampisNAK(rcvpkt)
udt_send(NAK)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
Wait for ACK or
NAK
Wait for call from
below
rdt_send(data)
Λ
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-29
Rdt20 Has a Fatal Flaw
What happens if ACKNAK corruptedSender doesnrsquot know what happened at receiverCanrsquot just retransmit possible duplicate
Handling duplicates Sender retransmits current pkt if ACKNAK garbledSender adds sequence number to each pktReceiver discards (doesnrsquot deliver up) duplicate pkt
Sender sends one packet then waits for receiver response
stop and wait
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-30
Sender whenever sender receives control message it sends a packet to receiver
A valid ACK Sends next packet (if exists) with new sequence A NAK or corrupt response resends old packet
Receiver sends ACKNAK to senderIf received packet is corrupt send NAKIf received packet is valid and has different sequence as prev packet send ACK and deliver new data upIf received packet is valid and has same sequence as prev packet ie is a retransmission of duplicate send ACK
Note ACKNAK do not contain sequence
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-31
Rdt21 Sender Handles Garbled ACKNAKs
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
Wait for ACK or NAK 0 udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isNAK(rcvpkt) )
Wait forcall 1 from
above
Wait for ACK or NAK 1
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt)
Λ
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-32
Rdt21 Receiver Handles Garbled ACKNAKs
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
Wait for 1 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq0(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq0(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp not corrupt(rcvpkt) ampamphas_seq1(rcvpkt)
sndpkt = make_pkt(ACK chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt)
sndpkt = make_pkt(NAK chksum)udt_send(sndpkt)
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-33
Rdt21 Discussion
SenderSeq added to pktTwo seq rsquos (01) will suffice WhyMust check if received ACKNAK corrupted Twice as many states
State must ldquorememberrdquowhether ldquocurrentrdquo pkt has 0 or 1 seq
ReceiverMust check if received packet is duplicate
State indicates whether 0 or 1 is expected pkt seq
Note receiver can notknow if its last ACKNAK received OK at sender
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-34
Rdt22 a NAK-free Protocol
Same functionality as rdt21 using ACKs onlyInstead of NAK receiver sends ACK for last pkt received OK
Receiver must explicitly include seq of pkt being ACKed
Duplicate ACK at sender results in same action as NAK retransmit current pkt
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-35
Rdt22 Sender Receiver Fragments
Wait for call 0 from
above
sndpkt = make_pkt(0 data checksum)udt_send(sndpkt)
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||
isACK(rcvpkt1) )Wait for ACK
0sender FSM
fragment
Wait for 0 from below
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp has_seq1(rcvpkt)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(ACK1 chksum)udt_send(sndpkt)
rdt_rcv(rcvpkt) ampamp (corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)receiver FSM
fragment
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)
Λ
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-36
Rdt30 Channels with Errors and Loss
New assumptionUnderlying channel can also lose packets (data or ACKs)
Checksum seq ACKs retransmissions will be of help but not enough
Q how to deal with loss
Sender waits until certain data or ACK lost then retransmitsAny drawbacks
Approach sender waits ldquoreasonablerdquo amount of time for ACK Retransmits if no ACK received in this timeIf pkt (or ACK) just delayed (not lost)
Retransmission will be duplicate but use of seq rsquos already handles thisReceiver must specify seq of pkt being ACKed
Requires countdown timer
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-37
Rdt30 Sendersndpkt = make_pkt(0 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
Wait for
ACK0
Wait for call 1 from
above
sndpkt = make_pkt(1 data checksum)udt_send(sndpkt)start_timer
rdt_send(data)
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt0)stop_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt) ampamp isACK(rcvpkt1)stop_timer
udt_send(sndpkt)start_timer
timeout
udt_send(sndpkt)start_timer
timeout
Wait for call 0 from
above
Wait for
ACK1
Λrdt_rcv(rcvpkt)
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt1) )
Λrdt_rcv(rcvpkt)Λ
rdt_rcv(rcvpkt) ampamp ( corrupt(rcvpkt) ||isACK(rcvpkt0) )
Λ
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-38
Rdt30 In Action
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-39
Rdt30 In Action
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-40
Performance of rdt30
Rdt30 works but performance stinksExample 1 Gbps link 15 ms e-e prop delay 1KB packet
Ttransmit = 8kbpkt109 bsec = 8 microsec
U sender utilization ndash fraction of time sender busy sending
U sender =
00830008
= 000027 L R RTT + L R
=
L (packet length in bits)R (transmission rate bps) =
1KB pkt every 30 msec -gt 33kBsec thruput over 1 Gbps linkNetwork protocol limits use of physical resources
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-41
Rdt30 Stop-and-Wait Operation
first packet bit transmitted t = 0
sender receiver
RTT
last packet bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
U sender =
00830008
= 000027 L R RTT + L R
=
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-42
Pipelined ProtocolsPipelining sender allows multiple ldquoin-flightrdquo
yet-to-be-acknowledged pktsRange of sequence numbers must be increasedBuffering at sender andor receiver
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-43
Pipelined Protocols
Advantage much better bandwidth utilization than stop-and-waitDisadvantage More complicated to deal with reliability issues eg corrupted lost out of order data
Two generic approaches to solving thisbull Go-Back-N protocolsbull Selective repeat protocols
Note TCP is not exactly either
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-44
Pipelining Increased Utilization
first packet bit transmitted t = 0sender receiver
RTT
last bit transmitted t = L R
first packet bit arriveslast packet bit arrives send ACK
ACK arrives send next packet t = RTT + L R
last bit of 2nd packet arrives send ACKlast bit of 3rd packet arrives send ACK
U sender =
024 30008
= 00008 3 L R RTT + L R
=
Increase utilizationby a factor of 3
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-45
Go-Back-NSender
K-bit seq in pkt headerldquoWindowrdquo of up to N consecutive unackrsquoed pkts allowed
ACK(n) ACKs all pkts up to including seq n - ldquocumulative ACKrdquoMay receive duplicate ACKs (see receiver)
Timer for each in-flight pktTimeout(n) retransmit pkt n and all higher seq pkts in windowCalled a sliding-window protocol
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-46
GBN Sender
rdt_Send() called checks to see if window is full No send out packetYes return data to application level
Receipt of ACK(n) cumulative acknowledgement that all packets up to and including n have been received Updates window accordingly and restarts timer
Timeout resends ALL packets that have been sent but not yet acknowledged
This is only event that triggers resend
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-47
GBN Sender Extended FSM
Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])hellipudt_send(sndpkt[nextseqnum-1])
timeout
rdt_send(data)
if (nextseqnum lt base+N) sndpkt[nextseqnum] = make_pkt(nextseqnumdatachksum)udt_send(sndpkt[nextseqnum])if (base == nextseqnum)
start_timernextseqnum++
elserefuse_data(data)
base = getacknum(rcvpkt)+1If (base == nextseqnum)
stop_timerelse
start_timer
rdt_rcv(rcvpkt) ampamp notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) ampamp corrupt(rcvpkt)
base=1nextseqnum=1
Λ
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-48
GBN Receiver Extended FSM
If expected packet receivedSend ACK and deliver packet upstairs
If out-of-order packet received Discard (donrsquot buffer) -gt no receiver bufferingRe-ACK pkt with highest in-order seq May generate duplicate ACKs
Wait
udt_send(sndpkt)default
rdt_rcv(rcvpkt)ampamp notcurrupt(rcvpkt)ampamp hasseqnum(rcvpktexpectedseqnum)
extract(rcvpktdata)deliver_data(data)sndpkt = make_pkt(expectedseqnumACKchksum)udt_send(sndpkt)expectedseqnum++
expectedseqnum=1sndpkt =
make_pkt(expectedseqnumACKchksum)
Λ
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-49
More on Receiver
ACK-only the receiver always sends ACK for last correctly received packet with highest in-order seq Receiver only sends ACKS (no NAKs)Need only remember expectedseqnumCan generate duplicate ACKs
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-50
GBN InAction
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-51
GBN is easy to code but might have performance problems
In particular if many packets are in pipeline at one time (bandwidth-delay product large) then one error can force retransmission of huge amounts of data
Selective Repeat protocol allows receiver to buffer data and only forces retransmission of required packets
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-52
Selective Repeat
Receiver individually acknowledges all correctly received pkts
Buffers pkts as needed for eventual in-order delivery to upper layer
Sender only resends pkts for which ACK not received
Sender timer for each unACKed pktCompare to GBN which only had timer for base packet
Sender windowN consecutive seq rsquosAgain limits seq s of sent unACKed pktsImportant Window size lt seq range
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-53
Selective Repeat Sender Receiver Windows
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-54
Selective Repeat
Data from above If next available seq in window send pkt
Timeout(n)Resend pkt n restart timer
ACK(n) in [sendbasesendbase+N]
Mark pkt n as receivedIf n smallest unACKed pkt advance window base to next unACKed seq
senderpkt n in [rcvbase rcvbase+N-1]
Send ACK(n)Out-of-order bufferIn-order deliver (also deliver buffered in-order pkts) advance window to next not-yet-received pkt
pkt n in [rcvbase-Nrcvbase-1]
ACK(n)Otherwise
Ignore
receiver
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-55
Selective Repeat In Action
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-56
Selective RepeatDilemma
Example seq rsquos 0 1 2 3window size=3
Receiver sees no difference in two scenariosIncorrectly passes duplicate data as new in (a)
Q what relationship between seq size and window size
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-57
GBN vs Selective Repeat
Selective repeat is more complicated as it needs buffering at the receiver but only retransmit packets required for retransmission
GBN is simpler but can lead to large number of unnecessary retransmission
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-58
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-59
TCP Overview RFCs 793 1122 1323 2018 2581
Full duplex dataBi-directional data flow in same connectionMSS maximum segment size
Connection-orientedHandshaking (exchange of control msgs) initrsquos sender receiver state before data exchange
Flow controlledSender will not overwhelm receiver
Point-to-pointOne sender one receiver
Reliable in-order byte steam
No ldquomessage boundariesrdquoPipelined
TCP congestion and flow control set window size
Send amp receive buffers
socketdoor
TCPsend buffer
TCPreceive buffer
socketdoor
segment
applicationwrites data
applicationreads data
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-60
More TCP DetailsMaximum Segment Size (MSS)
Depends upon implementation (can often be set)The Max amount of application-layer data in segment
Application Data + TCP Header = TCP Segment
Three way HandshakeClient sends special TCP segment to server requesting connection No payload (Application data) in this segmentServer responds with second special TCP segment
(again no payload)Client responds with third special segment
This can contain payload
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-61
More TCP Details (cont)
A TCP connection between client and server creates in both client and server
(i) Buffers(ii) Variables and
(iii) A socket connection to process
TCP only exists in the two end machinesNo buffers and variables allocated to the connection in any of the network elements between the host and server
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-62
TCP Segment Structure
source port dest port
32 bits
applicationdata
(variable length)
sequence numberacknowledgement number
Receive windowUrg data pnterchecksum
FSRPAUheadlen
notused
Options (variable length)
URG urgent data (generally not used)
ACK ACK valid
PSH push data now(generally not used)
RST SYN FINconnection estab(setup teardown
commands)
bytes rcvr willingto accept
countingby bytes of data(not segments)
Internetchecksum
(as in UDP)
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-63
TCP Seq rsquos and ACKsSeq rsquos
Byte stream ldquonumberrdquo of first byte in segmentrsquos data
ACKsSeq of next byte expected from other sideCumulative ACK TCP only acknowledge bytes up to the first missing byte in the stream
Host A Host B
Bytes [0~535]
Seq=XX1 ACK=536 data = lsquoYYYYrsquohellip
hellip
Bytes [0~535]
[9001000]Seq=XXN ACK=536 data = lsquoZZZZrsquo
Q how receiver handles out-of-order segmentsA TCP spec doesnrsquot say - up to implementer
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-64
TCP Seq rsquos and ACKs
Host A Host B
Seq=42 ACK=79 data = lsquoCrsquo
Seq=79 ACK=43 data = lsquoCrsquo
Seq=43 ACK=80
Usertypes
lsquoCrsquo
host ACKsreceipt
of echoedlsquoCrsquo
host ACKsreceipt oflsquoCrsquo echoes
back lsquoCrsquo
time
Simple telnet scenario
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-65
TCP Round Trip Time and Timeout
Q how to set TCP timeout valueLonger than RTT
But RTT variesToo short premature timeout
Unnecessary retransmissions
Too long slow reaction to segment loss
Q how to estimate RTTSampleRTT measured time from segment transmission until ACK receipt
Ignore retransmissions
SampleRTT will vary want estimated RTT ldquosmootherrdquo
Average several recent measurements not just current SampleRTT
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-66
TCP Round Trip Time and Timeout
EstimatedRTT= (1- α)EstimatedRTT + αSampleRTT
Exponential weighted moving averageInfluence of past sample decreases exponentially fastTypical value α = 0125
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-67
Example RTT EstimationRTT gaiacsumassedu to fantasiaeurecomfr
100
150
200
250
300
350
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
RTT
(mill
iseco
nds)
SampleRTT Estimated RTT
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-68
TCP Round Trip Time and Timeout
Setting the timeoutEstimtedRTT plus ldquosafety marginrdquo
Large variation in EstimatedRTT -gt larger safety margin
First estimate of how much SampleRTT deviates from EstimatedRTT
TimeoutInterval = EstimatedRTT + 4DevRTT
DevRTT = (1-β)DevRTT +β|SampleRTT-EstimatedRTT|
(typically β = 025)
Then set timeout interval
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-69
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-70
TCP Reliable Data Transfer
TCP creates rdtservice on top of IPrsquos unreliable servicePipelined segmentsCumulative acksTCP uses single retransmission timer
Retransmissions are triggered by
Timeout eventsDuplicate acks
Initially consider simplified TCP sender
Ignore duplicate acksIgnore flow control congestion control
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-71
TCP Sender EventsData rcvd from app
Create segment with seq Seq is byte-stream number of first data byte in segmentStart timer if not already running (think of timer as for oldest unacked segment)Expiration interval TimeOutInterval
TimeoutRetransmit segment that caused timeoutRestart timer
Ack rcvdIf acknowledges previously unackedsegments
Update what is known to be ackedStart timer if there are outstanding segments
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-72
TCP Sender(simplified)
NextSeqNum = InitialSeqNumSendBase = InitialSeqNum
loop (forever) switch(event)
event data received from application above create TCP segment with sequence number NextSeqNumif (timer currently not running)
start timerpass segment to IP NextSeqNum = NextSeqNum + length(data)
event timer timeoutretransmit not-yet-acknowledged segment with
smallest sequence numberstart timer
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
end of loop forever
Commentbull SendBase-1 last cumulatively ackrsquoed byte
Examplebull SendBase-1 = 71y= 73 so the rcvrwants 73+ y gt SendBase sothat new data is acked
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-73
TCP Retransmission ScenariosHost A
Seq=100 20 bytes data
ACK=100
time premature timeout
Host B
Seq=92 8 bytes data
ACK=120
Seq=92 8 bytes data
Seq=
92 t
imeo
ut
ACK=120
Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
lost ACK scenario
Host B
X
Seq=92 8 bytes data
ACK=100
timeSe
q=92
tim
eout
SendBase= 100
SendBase= 120
SendBase= 120
Sendbase= 100
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-74
TCP Retransmission Scenarios (more)Host A
Seq=92 8 bytes data
ACK=100
loss
tim
eout
Cumulative ACK scenario
Host B
X
Seq=100 20 bytes data
ACK=120
time
SendBase= 120
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-75
TCP ACK Generation [RFC 1122 RFC 2581]
Event at Receiver
Arrival of in-order segment withexpected seq All data up toexpected seq already ACKed
Arrival of in-order segment withexpected seq One other segment has ACK pending
Arrival of out-of-order segmenthigher-than-expect seq Gap detected
Arrival of segment that partially or completely fills gap
TCP Receiver action
Delayed ACK Wait up to 500msfor next segment If no next segmentsend ACK
Immediately send single cumulative ACK ACKing both in-order segments
Immediately send duplicate ACK indicating seq of next expected byte
Immediate send ACK provided thatsegment startsat lower end of gap
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-76
More on Sender Policies
Doubling the Timeout IntervalUsed by most TCP implementationsIf timeout occurs then after retransmisison Timeout Interval is doubledIntervals grow exponentially with each consecutive timeoutWhen Timer restarted because of (i) new data from above or (ii) ACK received then Timeout Interval is reset as described previously using Estimated RTT and DevRTTLimited form of Congestion Control
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-77
Fast Retransmit
Time-out period often relatively long
Long delay before resending lost packet
Detect lost segments via duplicate ACKs
Sender often sends many segments back-to-backIf segment is lost there will likely be many duplicate ACKs
If sender receives 3 ACKs for the same data it supposes that segment after ACKeddata was lost
Fast retransmit resend segment before timer expires
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-78
event ACK received with ACK field value of y if (y gt SendBase)
SendBase = yif (there are currently not-yet-acknowledged segments)
start timer
else increment count of dup ACKs received for yif (count of dup ACKs received for y = 3)
resend segment with sequence number y
Fast Retransmit Algorithm
a duplicate ACK for already ACKed segment
fast retransmit
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-79
TCP GBN or Selective Repeat
Basic TCP looks a lot like GBN
Many TCP implementations will buffer received out-of-order segments and then ACK them all after filling in the range
This looks a lot like Selective Repeat
TCP is a hybrid
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-80
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-81
TCP Flow Control
Sender should not overwhelm receiverrsquos capacity to receive dataIf necessary sender should slow down transmission rate to accommodate receiverrsquos rateDifferent from Congestion Controlwhose purpose was to handle congestion in networkBut both congestion control and flow control work by slowing down data transmission
low capacity receiver
fast network
flow control problem
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-82
TCP Flow Control
Receive side of TCP connection has a receive buffer
Speed-matching service matching the send rate to the receiving apprsquos drain rateApp process may be
slow at reading from buffer
sender wonrsquot overflowreceiverrsquos buffer by
transmitting too muchtoo fast
flow control
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-83
TCP Flow Control How it Works
(Suppose TCP receiver discards out-of-order segments)Spare room in buffer
= RcvWindow= RcvBuffer-[LastByteRcvd -
LastByteRead]
Rcvr advertises spare room by including value of RcvWindow in segmentsSender limits unACKeddata to RcvWindow
Guarantees receive buffer doesnrsquot overflow
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-84
Technical IssueSuppose RcvWindow=0 and that receiver has already ACKrsquod ALL packets in bufferSender does not transmit new packets until it hears RcvWindowgt0Receiver never sends RcvWindowgt0 since it has no new ACKS to send to SenderDEADLOCK
Solution TCP specs require sender to continue sending packets with one data byte while RcvWindow=0 just to keep receiving ACKS from receiverAt some point the receiverrsquos buffer will empty and RcvWindowgt0 will be transmitted back to sender
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-85
Note on UDP
UDP has no flow control
UDP appends packets to receiving socketrsquos buffer If buffer is full then packets are lost
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-86
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-87
TCP Connection Management
Recall TCP sender receiver establish ldquoconnectionrdquo before exchanging data segmentsInitialize TCP variables
Seq sBuffers flow control info (eg RcvWindow)
Client connection initiatorSocket clientSocket
= new Socket(hostname port number)
Server contacted by clientSocket connectionSocket
= welcomeSocketaccept()
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-88
TCP Connection Management (Cont)
Three way handshakeStep 1 client host sends TCP SYN
segment to serverSpecifies client initial seq No application data
Step 2 server host receives SYN replies with SYNACK segment
Server allocates buffersSpecifies server initial seq
Step 3 client receives SYNACK replies with ACK segment which may contain data
client
Connection request (SYN=1 seq=client_isn)
server
Connection granted (SYN=1 server_isn
ACK (SYN=0 seq=client_isn+1)
ack=client_isn+1)
ack=server_isn+1
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-89
TCP Connection Management (cont)
Closing a connection
client closes socketclientSocketclose()
Step 1 client end system sends TCP FIN control segment to server
Step 2 server receives FIN replies with ACK Closes connection sends FIN
client
FIN
server
ACK
FIN
close
close
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-90
TCP Connection Management (cont)
Step 3 client receives FIN replies with ACK
Enters ldquotimed waitrdquo -will respond with ACK to received FINs
Step 4 server receives ACK Connection closed
Note with small modification can handle simultaneous FINs
client
FIN
server
ACK
ACK
FIN
close
closing
closedti
med
wai
tclosed
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-91
TCP Connection Management (cont)
TCP client lifecycle
TCP server lifecycle
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-92
A few Special Cases
Have not discussed what happens if both client and server decide to close down connection at same time
It is possible that first ACK (from server) and second FIN (also from server) are sent in same segment
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-93
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-94
Principles of Congestion Control
CongestionInformally ldquotoo many sources sending too much data too fast for network to handlerdquoDifferent from flow controlManifestations
Lost packets (buffer overflow at routers)Long delays (queueing in router buffers)
A top-10 problem
high capacity receiver
congested network
congestion control problem
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-95
CausesCosts of Congestion
Examine 3 scenarios in which congestion occursLook at
Why congestion occursThe cost of congestion (resource utilization and performance at the end system)
Not focus onHow to react to or avoid congestion
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-96
CausesCosts of Congestion Scenario 1
Two senders two receiversOne router infinite buffers No retransmission
Large delays when congestedMaximum achievable throughput
unlimited shared output link buffers
Host Aλin original data
Host B
λout
Throughput and delay as a function of host sending rate
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-97
CausesCosts of Congestion Scenario 2
One router finite buffers Sender retransmission of lost packet
finite shared output link buffers
Host A λin original data
Host B
λout
λin original data plus retransmitted data
Offered Load
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-98
CausesCosts of Congestion Scenario 2
Always λin = λout (goodput)Case (a) magic transmission
Only send when therersquos space in buffer and no packet lossCase (b) ldquoperfectrdquo retransmission only when loss λrsquoin gt λout
Case (c) retransmission of delayed (not lost) packet makes λout
larger (than perfect case) for same λrsquoin
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-99
CausesCosts of Congestion Scenario 2
Brief summary of the ldquocostsrdquo of congestion so farLarge queuing delays as the packet-arrival rate nears the link capacitySender perform retransmissions to compensate the lost packets due to buffer overflowUnneeded retransmissions in the face of large delays
Link carries multiple copies of pkt
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-100
CausesCosts of Congestion Scenario 3 Four sendersMultihop pathsTimeoutretransmit
λin
Q what happens as and increase λ
in
finite shared output link buffers
Host Aλin original data
Host B
λout
λin original data plus retransmitted data
Host C
Host D
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-101
CausesCosts of Congestion Scenario 3
Another ldquocostrdquo of congestionWhen packet dropped any ldquoupstream transmission capacityrdquo used for that packet was wasted
Host A
Host B
λou
t
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-102
Approaches towards Congestion Control
End-end congestion controlNo explicit feedback from networkCongestion inferred from end-system observed loss delayApproach taken by TCP
Network-assisted congestion controlRouters provide feedback to end systems
Single bit indicating congestion (SNA DECbit TCPIP ECN ATM)Explicit rate sender should send at
Two broad approaches towards congestion control
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-103
Two Ways for Providing Feedback
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-104
Case study ATM ABR Congestion Control
ABR available bit rateldquoElastic servicerdquoIf senderrsquos path ldquounderloadedrdquo
Sender should use available bandwidth
If senderrsquos path congested
Sender throttled to minimum guaranteed rate
RM (resource management) cellsSent by sender interspersed with data cellsBits in RM cell set by switches (ldquonetwork-assistedrdquo)
NI bit no increase in rate (mild congestion)CI bit congestion indication
RM cells returned to sender by receiver with bits intact
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-105
Case study ATM ABR Congestion Control
Two-byte ER (explicit rate) field in RM cellCongested switch may lower ER value in cellSenderrsquo send rate thus maximum supportable rate on path
EFCI bit in data cells set to 1 in congested switchIf data cell preceding RM cell has EFCI set sender sets CI bit in returned RM cell
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-106
Chapter 3 Outline
31 Transport-layer services32 Multiplexing and demultiplexing33 Connectionless transport UDP34 Principles of reliable data transfer
35 Connection-oriented transport TCP
Segment structureReliable data transferFlow controlConnection management
36 Principles of congestion control37 TCP congestion control
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-107
TCP Congestion ControlEnd-to-end control (no network assistance)Transmission rate limited by congestion window size Congwin over segments
Congwin dynamically modified to reflect perceived congestion
w segments each with MSS bytes sent in one RTT
throughput = w MSSRTT Bytessec
Congwin
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-108
To simplify presentation we assume that RcvBuffer is large enough that it will not overflow Tools are ldquosimilarrdquo to flow controlsender limits transmission usingLastByteSent-LastByteAcked le CongWin
How does sender perceive congestionloss event = timeout or 3 duplicate ACKsTCP sender reduces rate (CongWin) after loss event
CongWin incrementACK arrives at slow rate -gt CongWin be increased at a slow rateACK arrives at high rate -gt CongWin be increased at a high rate
Self-ClockingTCP uses ACK to trigger (clock) its increase in congwin
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-109
Three mechanismsAIMD = Additive Increase Multiplicative DecreaseSlow start = CongWin set to 1 and then grows exponentially
Conservative after timeout events
TCP Congestion Control Algorithms
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-110
TCP AIMD
8 Kbytes
16 Kbytes
24 Kbytes
time
congestionwindow
Multiplicative decreasecut CongWin in half after loss event
Additive increase increase CongWin by 1 MSS every RTT in the absence of loss events probing also known as congestion avoidance
Long-lived TCP connection
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-111
TCP Slow Start
When connection begins CongWin = 1 MSS
Example MSS = 500 bytes amp RTT = 200 msecInitial rate = 20 kbps
Available bandwidth may be gtgt MSSRTT
Desirable to quickly ramp up to respectable rate
When connection begins increase rate exponentially fast until first loss event
Slow Start is NOT really Slow
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-112
TCP Slow Start (more)
When connection begins increase rate exponentially until first loss event
Double CongWin every RTTDone by incrementing CongWin for every ACK received
Summary initial rate is slow but ramps up exponentially fast
one segment
RTT
two segments
Host A Host B
time
four segments
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-113
So FarSlow-Start ramps up exponentiallyFollowed by AIMD sawtooth patternQ When should the exponential increase switch to linear
Reality (TCP Reno)Introduce new variable threshold initially very largeSlow-Start exponential growth stops when reaches threshold and then switches to AIMDTwo different types of loss events
bull 3 dup ACKS cut CongWin in half and set threshold=CongWin (now in standard AIMD)
bull Timeout set threshold=CongWin2 CongWin=1 and switch to Slow-Start
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-114
Reason for treating 3 dup ACKS differently than timeout is that
3 dup ACKs indicates network capable of delivering some segments while Timeout before 3 dup ACKs is ldquomore alarmingrdquo
Note that older protocol TCP Tahoe treated both types of loss events the same and always goes to slow-start with Congwin=1 after a loss event
TCP Renorsquos skipping of the slow start for a 3-DUP-ACK loss event is known as fast-recovery
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-115
Summary TCP Congestion Control
When CongWin is below Threshold sender in slow-start phase window grows exponentially
When CongWin is above Threshold sender is in congestion-avoidance phase window grows linearly
When a triple duplicate ACK occurs Thresholdset to CongWin2 and CongWin set to Threshold(only in TCP Reno)
When timeout occurs Threshold set to CongWin2 and CongWin is set to 1 MSS (TCP Tahoe does this for 3 Dup Acks as well)
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-116
The Big Picture
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-117
TCP Sender Congestion Control
Duplicate ACK
Timeout
Loss event detected by triple duplicate ACK
ACK receipt for previously unacked data
ACK receipt for previously unacked data
Event
SS or CA
SS or CA
SS or CA
CongestionAvoidance (CA)
Slow Start (SS)
State
CongWin and Threshold not changed
Increment duplicate ACK count for segment being acked
Enter slow startThreshold = CongWin2 CongWin = 1 MSSSet state to ldquoSlow Startrdquo
Fast recovery implementing multiplicative decrease CongWin will not drop below 1 MSS
Threshold = CongWin2 CongWin = ThresholdSet state to ldquoCongestion Avoidancerdquo
Additive increase resulting in increase of CongWin by 1 MSS every RTT
CongWin = CongWin+MSS (MSSCongWin)
Resulting in a doubling of CongWin every RTT
CongWin = CongWin + MSS If (CongWin gt Threshold)
set state to ldquoCongestion Avoidancerdquo
CommentaryTCP Sender Action
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-118
TCP Throughput
Whatrsquos the average throughout of TCP as a function of window size and RTT
Ignore slow startLet W be the window size when loss occursWhen window is W throughput is WRTTJust after loss window drops to W2 throughput to W2RTTAverage throughout 75 WRTT
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-119
TCP Futures TCP over ldquoLong Fat Pipesrdquo
Example 1500 byte segments 100ms RTT want 10 Gbps throughputRequires window size W = 83333 in-flight segmentsThroughput in terms of loss rate
L = 210-10 Wow
New versions of TCP for high-speed needed
LRTTMSSsdot221
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-120
Fairness goal if K TCP sessions share same bottleneck link of bandwidth R each should have average rate of RK
TCP Fairness
TCP connection 1
bottleneckrouter
capacity RTCP connection 2
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-121
Why is TCP fairTwo competing sessions
Additive increase gives slope of 1 as throughout increasesmultiplicative decrease decreases throughput proportionally
R
R
equal bandwidth share
Connection 1 throughput
Conn
ecti
on 2
thr
ough
p ut
congestion avoidance additive increaseloss decrease window by factor of 2
congestion avoidance additive increaseloss decrease window by factor of 2
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-122
Fairness (more)Fairness and UDP
Multimedia apps often do not use TCP
Do not want rate throttled by congestion control
Instead use UDPpump audiovideo at constant rate tolerate packet loss
Research area TCP friendly
Fairness and parallel TCP connectionsNothing prevents app from opening parallel connections between 2 hostsWeb browsers do this Example link of rate R supporting 9 cnctions
New app asks for 1 TCP gets rate R10New app asks for 11 TCPs gets R2
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo
Transport Layer 3-123
Chapter 3 SummaryPrinciples behind transport layer services
Multiplexing demultiplexingReliable data transferFlow controlCongestion control
Instantiation and implementation in the Internet
UDPTCP
NextLeaving the network ldquoedgerdquo (application transport layers)Into the network ldquocorerdquo