Download - TCP Congestion Control
![Page 1: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/1.jpg)
TCPCongestion Control
Computer Networks TCP Congestion Control 1
![Page 2: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/2.jpg)
Principles of Congestion Control
Congestion::• informally: “too many sources sending too much data
too fast for the network to handle”• different from flow control!• manifestations:
– lost packets (buffer overflow at routers)– long delays (queueing in router buffers)
• a major problem in networking!
Computer Networks TCP Congestion Control 2
![Page 3: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/3.jpg)
Causes/Costs of CongestionScenario 1
• two senders, two receivers
• one router, infinite buffers
• no retransmission
• large delays when congested
• maximum achievable throughput
unlimited shared output link buffers
Host Alin : original data
Host B
lout
Computer Networks TCP Congestion Control 3
![Page 4: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/4.jpg)
Causes/Costs of CongestionScenario 2
• one router, finite buffers • sender retransmits lost packets
finite shared output link buffers
Host A lin : original data
Host B
lout
l'in : original data, plus retransmitted data
Computer Networks TCP Congestion Control 4
offered load
![Page 5: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/5.jpg)
• always: (goodput)• “perfect” retransmission only when loss:• retransmission of delayed (not lost) packet makes larger (than
perfect case) for same
lin
lout=
lin
lout>
linlout
“costs” of congestion: more work (retransmissions) for a given “goodput” unneeded retransmissions: link carries multiple copies of packet
R/2
R/2lin
l out
b.
R/2
R/2lin
l out
a.
R/2
R/2lin
l out
c.
R/4
R/3
Causes/Costs of CongestionScenario 2
Computer Networks TCP Congestion Control 5
![Page 6: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/6.jpg)
Approaches towards Congestion Control
end-end congestion control:
• no explicit feedback from network
• congestion inferred from end-system observed loss, delay
• approach taken by TCP
network-assisted congestion control:
• routers provide feedback to end systems– single bit indicating
congestion (SNA, DECbit, TCP/IP ECN, ATM)
– explicit rate sender should use for sending.
Two broad approaches towards congestion control:
Computer Networks TCP Congestion Control 6
![Page 7: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/7.jpg)
TCP Congestion Control
Lecture material taken from “Computer Networks A Systems Approach”,
Fourth Edition,Peterson and Davie,Morgan Kaufmann, 2007.
Computer Networks TCP Congestion Control 7
![Page 8: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/8.jpg)
TCP Congestion Control• Essential strategy :: The TCP host sends
packets into the network without a reservation and then the host reacts to observable events.
• Originally TCP assumed FIFO queuing.• Basic idea :: each source determines how
much capacity is available to a given flow in the network.
• ACKs are used to ‘pace’ the transmission of packets such that TCP is “self-clocking”.
Computer Networks TCP Congestion Control 8
![Page 9: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/9.jpg)
TCP Congestion Control• Goal: TCP sender should transmit as fast as
possible, but without congesting network.• issue - how to find rate just below congestion
level?• Each TCP sender sets its window size, based
on implicit feedback: • ACK segment received network is not
congested, so increase sending rate.• lost segment - assume loss due to
congestion, so decrease sending rate.
K & R
Computer Networks TCP Congestion Control 9
![Page 10: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/10.jpg)
TCP Congestion Control• “probing for bandwidth”: increase transmission rate on
receipt of ACK, until eventually loss occurs, then decrease transmission rate
• continue to increase on ACK, decrease on loss (since available bandwidth is changing, depending on other connections in network).
ACKs being received, so increase rate
X
X
XX
Xloss, so decrease rate
send
ing
rate
time
• Q: how fast to increase/decrease?
TCP’s“sawtooth”
behavior
K & R
Computer Networks TCP Congestion Control 10
![Page 11: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/11.jpg)
AIMD(Additive Increase / Multiplicative
Decrease)• CongestionWindow (cwnd) is a variable held by
the TCP source for each connection.
• cwnd is set based on the perceived level of congestion. The Host receives implicit (packet drop) or explicit (packet mark) as an indication of internal congestion.
MaxWindow :: min (CongestionWindow , AdvertisedWindow)
EffectiveWindow = MaxWindow – (LastByteSent -LastByteAcked)
Computer Networks TCP Congestion Control 11
![Page 12: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/12.jpg)
Additive Increase (AI)• Additive Increase is a reaction to perceived available
capacity (referred to as congestion avoidance stage).• Frequently in the literature, additive increase is defined
by parameter α (where the default is α = 1).• Linear Increase :: For each “cwnd’s worth” of packets
sent, increase cwnd by 1 packet.• In practice, cwnd is incremented fractionally for each
arriving ACK.
increment = MSS x (MSS /cwnd)cwnd = cwnd + increment
Computer Networks TCP Congestion Control 12
![Page 13: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/13.jpg)
Figure 6.8 Additive Increase
Source Destination
Add one packeteach RTT
Computer Networks TCP Congestion Control 13
![Page 14: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/14.jpg)
Multiplicative Decrease (MD)* Key assumption :: a dropped packet and resultant
timeout are due to congestion at a router.• Frequently in the literature, multiplicative decrease
is defined by parameter β (where the default is β = 0.5)
Multiplicate Decrease:: TCP reacts to a timeout by halving cwnd.
• Although defined in bytes, the literature often discusses cwnd in terms of packets (or more formally in MSS == Maximum Segment Size).
• cwnd is not allowed below the size of a single packet.
Computer Networks TCP Congestion Control 14
![Page 15: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/15.jpg)
AIMD(Additive Increase / Multiplicative
Decrease)• It has been shown that AIMD is a necessary
condition for TCP congestion control to be stable.• Because the simple CC mechanism involves
timeouts that cause retransmissions, it is important that hosts have an accurate timeout mechanism.
• Timeouts set as a function of average RTT and standard deviation of RTT.
• However, TCP hosts only sample round-trip time once per RTT using coarse-grained clock.
Computer Networks TCP Congestion Control 15
![Page 16: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/16.jpg)
Figure 6.9 Typical TCPSawtooth Pattern
60
20
1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0Time (seconds)
70
304050
10
10.0
Computer Networks TCP Congestion Control 16
![Page 17: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/17.jpg)
Slow Start• Linear additive increase takes too long to
ramp up a new TCP connection from cold start.
• Beginning with TCP Tahoe, the slow start mechanism was added to provide an initial exponential increase in the size of cwnd.
Remember mechanism by: slow start prevents a slow start. Moreover, slow start is slower than sending a full advertised window’s worth of packets all at once.
Computer Networks TCP Congestion Control 17
![Page 18: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/18.jpg)
Slow Start• The source starts with cwnd = 1.• Every time an ACK arrives, cwnd is
incremented.cwnd is effectively doubled per RTT “epoch”.• Two slow start situations:
At the very beginning of a connection {cold start}. When the connection goes dead waiting for a
timeout to occur (i.e, when the advertized window goes to zero!)
Computer Networks TCP Congestion Control 18
![Page 19: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/19.jpg)
Figure 6.10 Slow Start
Source Destination
Slow StartAdd one packet
per ACK
Computer Networks TCP Congestion Control 19
![Page 20: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/20.jpg)
Slow Start• However, in the second case the source
has more information. The current value of cwnd can be saved as a congestion threshold.
• This is also known as the “slow start threshold” ssthresh.
Computer Networks TCP Congestion Control 20
![Page 21: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/21.jpg)
ssthresh
Computer Networks TCP Congestion Control 21
![Page 22: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/22.jpg)
Figure 6.11 Behavior of TCPCongestion Control
60
20
1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0Time (seconds)
70
304050
10
Computer Networks TCP Congestion Control 22
![Page 23: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/23.jpg)
Fast Retransmit• Coarse timeouts remained a problem, and Fast
retransmit was added with TCP Tahoe.• Since the receiver responds every time a packet
arrives, this implies the sender will see duplicate ACKs.
Basic Idea:: use duplicate ACKs to signal lost packet.
Fast RetransmitUpon receipt of three duplicate ACKs, the TCP Sender
retransmits the lost packet.
Computer Networks TCP Congestion Control 23
![Page 24: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/24.jpg)
Fast Retransmit• Generally, fast retransmit eliminates about half
the coarse-grain timeouts.• This yields roughly a 20% improvement in
throughput.• Note – fast retransmit does not eliminate all
the timeouts due to small window sizes at the source.
Computer Networks TCP Congestion Control 24
![Page 25: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/25.jpg)
Figure 6.12 Fast Retransmit
Packet 1Packet 2Packet 3Packet 4
Packet 5Packet 6
Retransmitpacket 3
ACK 1ACK 2
ACK 2ACK 2
ACK 6
ACK 2
Sender Receiver
Fast Retransmit
Based on threeduplicate ACKs
Computer Networks TCP Congestion Control 25
![Page 26: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/26.jpg)
Figure 6.13 TCP Fast Retransmit Trace
60
20
1.0 2.0 3.0 4.0 5.0 6.0 7.0Time (seconds)
70
304050
10
Computer Networks TCP Congestion Control 26
![Page 27: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/27.jpg)
Fast Recovery• Fast recovery was added with TCP Reno.• Basic idea:: When fast retransmit detects
three duplicate ACKs, start the recovery process from congestion avoidance region and use ACKs in the pipe to pace the sending of packets.
Fast RecoveryAfter Fast Retransmit, half cwnd and commence
recovery from this point using linear additive increase‘primed’ by left over ACKs in pipe.
Computer Networks TCP Congestion Control 27
![Page 28: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/28.jpg)
Modified Slow Start• With fast recovery, slow start only
occurs:–At cold start–After a coarse-grain timeout
• This is the difference between TCP Tahoe and TCP Reno!!
Computer Networks TCP Congestion Control 28
![Page 29: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/29.jpg)
Many TCP ‘flavors’• TCP New Reno• TCP SACK
– requires sender and receiver both to support TCP SACK.
– possible state machine is complex.• TCP Vegas
– adjusts window size based on difference between expected and actual RTT.
• TCP BIC TCP Cubic {used by Linux}• TCP Compound {used by Windows}
Computer Networks TCP Congestion Control 29
![Page 30: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/30.jpg)
TCP New Reno• Two problem scenarios with TCP Reno
– bursty losses, Reno cannot recover from bursts of 3+ losses.
– Packets arriving out-of-order can yield duplicate acks when in fact there is no loss.
• New Reno solution – try to determine the end of a burst loss.
Computer Networks TCP Congestion Control 30
![Page 31: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/31.jpg)
TCP New Reno• When duplicate ACKs trigger a
retransmission for a lost packet, sender remembers the highest packet sent from window before entering Fast Recovery in recover.
• Upon receiving an ACK,– if ACK < recover => partial ACK– If ACK ≥ recover => new ACK
Computer Networks TCP Congestion Control 31
![Page 32: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/32.jpg)
TCP New Reno• Partial ACK implies another lost
packet: retransmit next packet, inflate window and stay in fast recovery.
• New ACK implies fast recovery is over: starting from 0.5 x cwnd proceed with congestion avoidance (linear increase).
• New Reno recovers from n losses in n round trips.
Computer Networks TCP Congestion Control 32
![Page 33: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/33.jpg)
Figure 5.6 Three-way TCP Handshake
Computer Networks TCP Congestion Control 33
![Page 34: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/34.jpg)
Adaptive Retransmissions
RTT:: Round Trip Time between a pair of hosts on the Internet.
• The initial TCP TimeOut value (RIO) is an operating system parameter that is set very conservatively!
• How to set the TimeOut value (RTO)?– The timeout value is set as a function of
the expected RTT.– Consequences of a bad choice?
Computer Networks TCP Congestion Control 34
![Page 35: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/35.jpg)
Original Algorithm• Keep a running average of RTT and
compute TimeOut as a function of this RTT.– Send packet and keep timestamp ts .– When ACK arrives, record timestamp ta .
SampleRTT = ta - ts
Computer Networks TCP Congestion Control 35
![Page 36: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/36.jpg)
Original AlgorithmCompute a weighted average:
EstimatedRTT = α x EstimatedRTT + (1- α) x SampleRTT
Original TCP spec: α in range (0.8,0.9)
TimeOut = 2 x EstimatedRTT
Computer Networks TCP Congestion Control 36
![Page 37: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/37.jpg)
Karn/Partidge Algorithm
An obvious flaw in the original algorithm:
Whenever there is a retransmission it is impossible to know whether to associate the ACK with the original packet or the retransmitted packet.
Computer Networks TCP Congestion Control 37
![Page 38: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/38.jpg)
Figure 5.10 Associating the ACK?
Sender ReceiverOriginal transmission
ACK
Retransmission
Sender ReceiverOriginal transmission
ACKRetransmission
(a) (b)
Computer Networks TCP Congestion Control 38
![Page 39: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/39.jpg)
Karn/Partidge Algorithm
1. Do not measure SampleRTT when sending packet more than once.
2. For each retransmission, set TimeOut to double the last TimeOut.{ Note – this is a form of exponential backoff based on the believe that the lost packet is due to congestion.}
Computer Networks TCP Congestion Control 39
![Page 40: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/40.jpg)
Jacobson/Karels Algorithm
The problem with the original algorithm is that it did not take into account the variance of SampleRTT.
Difference = SampleRTT – EstimatedRTTEstimatedRTT = EstimatedRTT +
(δ x Difference)Deviation = δ (|Difference| - Deviation)
where δ is a fraction between 0 and 1.
Computer Networks TCP Congestion Control 40
![Page 41: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/41.jpg)
Jacobson/Karels Algorithm
TCP computes timeout using both the mean and variance of RTT
TimeOut = µ x EstimatedRTT + Φ x Deviation
where based on experience µ = 1 and Φ = 4.
Computer Networks TCP Congestion Control 41
![Page 42: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/42.jpg)
TCP Congestion Control Summary
• Congestion occurs due to a variety of circumstance.
• TCP interacts with routers in the subnet and reacts to implicit congestion notification (packet drop) by reducing the TCP sender’s congestion window (MD).
• TCP increases congestion window using slow start or congestion avoidance (AI).
Computer Networks TCP Congestion Control 42
![Page 43: TCP Congestion Control](https://reader035.vdocument.in/reader035/viewer/2022062510/5681672e550346895ddbd34b/html5/thumbnails/43.jpg)
TCP Congestion Control Summary• Important TCP Congestion Control ideas
include: AIMD, Slow Start, Fast Retransmit and Fast Recovery.
• Know the differences between TCP Tahoe, TCP Reno and TCP New Reno.
• Currently, the two most common versions of TCP are Compound (Windows) and Cubic (Linux).
• TCP needs rules and an algorithm to determine RIO and RTO.Computer Networks TCP Congestion Control 43