CH 4 THE MEDIUM ACCESS CONTROL SUBLAYER

1

Medium Access Control: a means of controlling access to the medium to promote orderly and efficient use.

OSI MODEL AND PROJECT 802

2

THE CHANNEL ALLOCATION PROBLEM

• Static Channel Allocation in LANs and MANsFDM: small and constant users, heavy load of traffic of each.TDM:same problem. Poor performance.None of the static channel allocation methods work well with bursty traffic.

• Dynamic Channel Allocation in LANs and MANs

3

PURE ALOHA

In pure ALOHA, frames are transmitted at completely arbitrary times.

4

Multiple Access Protocols

PURE ALOHA (2)

Vulnerable period for the shaded frame. 5

SLOTTED ALOHA Time in uniform slots equal to

frame transmission timeNeed central clock (or other sync

mechanism)Transmission begins at slot

boundaryFrames either miss or overlap

totallyMax utilization 36.8% 6

RELATIVE FORMULAS FOR THE ALOHA

7

Throughput orChannel Utilization

Probability of collision Probability of success

Pure ALOHASlotted ALOHA

tG GeGeS 22 tGr eecollisionP 22 11)(

tGr eeSuccessP 22)(

tG GeGeS tGr eecollisionP 11)( tG

r eeSuccessP )(

tG sec/: request slotrequestG /:

rate data

size frame

vilocity

distance fp ttt

PURE ALOHA AND SLOTTED ALOHA

Throughput versus offered traffic for ALOHA systems.8

PERSISTENT AND NONPERSISTENT CSMA

All stations know that a transmission has started almost immediately

First listen for clear medium (carrier sense)

If medium idle, transmit with a probability.

If two stations start at the same instant, collision

Propagation time is much less than transmission time

Wait reasonable time (round trip plus ACK contention)

No ACK then retransmit

9

PERSISTENT AND NONPERSISTENT CSMA

Comparison of the channel utilization versus load for various random access protocols.

10

CSMA/CD (WITH COLLISION DETECTION)

If collision detected, jam then cease transmission rather than finish transmitting their frame

After jam, wait random time then start again

Half-duplex systemSave time and bandwidth.Basis of Ethernet LAN.

11

CSMA/CDOPERATION

12

TOKEN RING (802.5)

MAC protocolSmall frame (token) circulates when idleStation waits for tokenChanges one bit in token to make it SOF for data

frameAppend rest of data frameFrame makes round trip and is absorbed by

transmitting stationStation then inserts new token when transmission

has finished and leading edge of returning frame arrives

Under light loads, some inefficiencyUnder heavy loads, round robin makes efficiency

and fair. 13

TOKEN RINGOPERATION

14

FDDI MAC PROTOCOL

Fiber Distributed Data InterfaceAs for 802.5 except:

Station seizes token by aborting token transmission

Once token captured, one or more data frames transmitted

New token released as soon as transmission finished

15

ETHERNET

• Ethernet Cabling• Manchester Encoding• The Ethernet MAC Sublayer Protocol• The Binary Exponential Backoff

Algorithm• Ethernet Performance• Switched Ethernet• Fast Ethernet• Gigabit Ethernet• IEEE 802.2: Logical Link Control

16

13.17

Ethernet evolution through four generations

13

.18

Figure 13.8 Categories of Standard Ethernet

ETHERNET CABLING

The most common kinds of Ethernet cabling. 19

13

.20

Figure 13.9 Encoding in a Standard Ethernet implementation

13

.21

Figure 13.10 10Base5 implementation

13

.22

Figure 13.11 10Base2 implementation

13

.23

Figure 13.12 10Base-T implementation

13

.24

Figure 13.13 10Base-F implementation

13

.25

Table 13.1 Summary of Standard Ethernet implementations

ETHERNET TOPOLOGY

Cable topologies. (a) Linear, (b) Spine, (c) Tree, (d) Segmented.

26

BASEBAND CONFIGURATION The size limitation is

usually solved by using repeaters to divide the medium into smaller segments

Repeaters relay digital signals in both directions, making the segments appear like one medium

As repeaters recover the digital signal, they remove any attenuation 27

13

.28

Figure 13.15 A network with and without a bridge

13

.29

Figure 13.16 Collision domains in an unbridged network and a bridged network

13

.30

Figure 13.17 Switched Ethernet

13

.31

Figure 13.18 Full-duplex switched Ethernet

13

.32

13-4 FAST ETHERNET13-4 FAST ETHERNET

Fast Ethernet was designed to compete with LAN Fast Ethernet was designed to compete with LAN protocols such as FDDI or Fiber Channel. IEEE protocols such as FDDI or Fiber Channel. IEEE created Fast Ethernet under the name 802.3u. Fast created Fast Ethernet under the name 802.3u. Fast Ethernet is backward-compatible with Standard Ethernet is backward-compatible with Standard Ethernet, but it can transmit data 10 times faster at a Ethernet, but it can transmit data 10 times faster at a rate of 100 Mbps. rate of 100 Mbps.

MAC SublayerPhysical Layer

Topics discussed in this section:Topics discussed in this section:

13

.33

Figure 13.19 Fast Ethernet topology

13

.34

Figure 13.20 Fast Ethernet implementations

13

.35

Figure 13.21 Encoding for Fast Ethernet implementation

13

.36Table 13.2 Summary of Fast Ethernet implementations

13

.37

13-5 GIGABIT ETHERNET13-5 GIGABIT ETHERNET

The need for an even higher data rate resulted in the The need for an even higher data rate resulted in the design of the Gigabit Ethernet protocol (1000 Mbps). design of the Gigabit Ethernet protocol (1000 Mbps). The IEEE committee calls the standard 802.3z.The IEEE committee calls the standard 802.3z.

MAC SublayerPhysical Layer

Ten-Gigabit Ethernet

Topics discussed in this section:Topics discussed in this section:

13

.38

In the full-duplex mode of Gigabit Ethernet, there is no collision;the maximum length of the cable is determined by the signal attenuation

in the cable.

Note

13

.39

Figure 13.22 Topologies of Gigabit Ethernet

13

.40

Figure 13.23 Gigabit Ethernet implementations

13

.41

Figure 13.24 Encoding in Gigabit Ethernet implementations

13

.42Table 13.3 Summary of Gigabit Ethernet implementations

13

.43

Table 13.4 Summary of Ten-Gigabit Ethernet implementations

SWITCHED ETHERNET

A simple example of switched Ethernet. 44

FAST ETHERNETFAST ETHERNET

Fast Ethernet was designed to compete with LAN Fast Ethernet was designed to compete with LAN protocols such as FDDI or Fiber Channel. IEEE protocols such as FDDI or Fiber Channel. IEEE created Fast Ethernet under the name 802.3u. Fast created Fast Ethernet under the name 802.3u. Fast Ethernet is backward-compatible with Standard Ethernet is backward-compatible with Standard Ethernet, but it can transmit data 10 times faster at a Ethernet, but it can transmit data 10 times faster at a rate of 100 Mbps. rate of 100 Mbps.

45

Fast Ethernet topology

46

Fast Ethernet implementations

47

13.48

GIGABIT ETHERNETGIGABIT ETHERNET

The need for an even higher data rate resulted in the The need for an even higher data rate resulted in the design of the Gigabit Ethernet protocol (1000 Mbps). design of the Gigabit Ethernet protocol (1000 Mbps). The IEEE committee calls the standard 802.3z.The IEEE committee calls the standard 802.3z.

GIGABIT ETHERNET

(a) A two-station Ethernet.(b) A multistation Ethernet.

49

GIGABIT ETHERNET (2)

50

Gigabit Ethernet - DifferencesCarrier extensionAt least 4096 bit-times long (512 for 10/100)Frame bursting extended to 200m.New coding

Summary of Ten-Gigabit Ethernet implementations

51

IEEE standard for LANs

52

IEEE 802.2: LOGICAL LINK CONTROL

(a) Position of LLC. (b) Protocol formats.

53

54

ETHERNET MAC SUBLAYER PROTOCOL

WCB/McGraw-Hill The McGraw-Hill Companies, Inc., 1998

55

PDU FORMAT

WCB/McGraw-Hill The McGraw-Hill Companies, Inc., 1998

MINIMUM AND MAXIMUM LENGTH

56

13.57

Example of an Ethernet address in hexadecimal notation

ETHERNET PERFORMANCE

Efficiency of Ethernet at 10 Mbps with 512-bit slot times.

58

13.59

Figure 13.2 HDLC frame compared with LLC and MAC frames

LAN TRANSMISSION TECHNOLOGIES

Ethernet 10 Mbit/s Token Ring 4/16 Mbit/s Fast Ethernet 100 Mbit/s FDDI 100 Mbit/s Gigabit Ethernet 1 Gbit/s ATM 25 Mbit/s to 2.4

Gbit/sOnly Ethernet versions are growing

60