Wired LANs: Ethernet - Erode Sengunthar Engineering … LANs: Ethernet Chapter 13 Dr. Gihan NAGUIB ... If the upper- layer packet is less than 46 bytes, ... network. Dr. Gihan NAGUIB

12/6/2010

1

Behrouz A. Forouzan “Data communication and Networking 1

Wired LANs: Ethernet

Chapter 13

Dr. Gihan NAGUIB

“Behrouz A. Forouzan “Data communication and Networking” 2

1313--1 1 IEEE STANDARDSIEEE STANDARDS

InIn 19851985,, thethe ComputerComputer SocietySociety ofof thethe IEEEIEEE startedstarted aa

project,project, calledcalled ProjectProject 802802,, toto setset standardsstandards toto enableenable

intercommunicationintercommunication amongamong equipmentequipment fromfrom aa

varietyvariety ofof manufacturersmanufacturers.. ProjectProject 802802 isis aa wayway ofof

specifyingspecifying functionsfunctions ofof thethe physicalphysical layerlayer andand thethe

datadata linklink layerlayer ofof majormajor LANLAN protocolsprotocols..

Data Link Layer

Physical Layer

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

Dr. Gihan NAGUIB

12/6/2010

2


IEEE standard for LANs

Dr. Gihan NAGUIB

IEEE divided the Data link layer into two sublayer :

upper layer :logical link control (LLC); flow and

error control.

Lower sublayer : Multiple access (MAC); media

access control.

Multiple access (MAC) for resolving access to the

shared media.

If channel is dedicated ( point to point) we do not need

the (MAC); sublayer.


IEEE standard for LANs

Dr. Gihan NAGUIB

Token passingCSMA/CD

12/6/2010

3


LLC (Logical link control) and MAC (Media Access Control)

In IEEE project 802, flow control , error control, and part of the

framing duties are collected into one sublayer called the logical

link control (LLC )

LLC provides one single data link control for all IEEE LANs.

IEEE project 802 has created a sublayer MAC that defines the

specific access method for each LAN. In contrast to the LLC, MAC

contains a number of distinct modules: each defines the access method

and the framing format specific to the corresponding LAN protocol

For example:

CSMA/CD as media access method for Ethernet LANs.

Token passing method for Token Ring and Token Bus LANs.

Framing is handled in both the LLC and MAC sublayer.

Dr. Gihan NAGUIB

“Behrouz A. Forouzan “Data communication and Networking” 6Dr. Gihan NAGUIB

Physical layer

Physical layer is dependent on the implementation

and type of the physical media used.

IEEE define detailed specifications for each LAN

implementation.

For example, although there is only one MAC

sublayer for Standard Ethernet( CSMA/CD), there is

a different physical layer specifications for each

Ethernet implementations.

12/6/2010

4


DescriptionName

EthernetIEEE 802.3

Token busIEEE 802.4

Token RingIEEE 802.5

Wireless LAN IEEE 802.11a/b/g/n

BluetoothIEEE 802.15.1

IEEE 802 Series of LAN Standards


ETHERNET IEEE ETHERNET IEEE 802802..33

It is the dominant LAN technology.

Cheap

First widely used LAN technology

Simpler and cheaper than token LANs

Kept up with speed race: 10, 100, 1000 Mbps

12/6/2010

5


ETHERNET EvolutionETHERNET Evolution

TheThe originaloriginal EthernetEthernet waswas createdcreated inin 19761976 atat Xerox’sXerox’s

PaloPalo AltoAlto ResearchResearch CenterCenter (PARC)(PARC).. SinceSince then,then, itit hashas

gonegone throughthrough fourfour generationsgenerations..


ETHERNETETHERNET

The MAC sublayer governs the operation of the random

access method

Standard Ethernet uses CSMA/CD with 1-persistent

Ethernet dose not provide any mechanism for

acknowledging received frames( unreliable medium).

Acknowledgments must be implemented at the higher layer.

It also frames data received from the upper layer and

passes them to the physical layer.

MAC Sublayer

12/6/2010

6


Ethernet (IEEE 802.3) MAC frame

Preamble: 7bytes (56 bits); Alternating 0s and 1s, used for synchronizing

Start Frame Delimiter (SFD):

10101011 indicates the start of the frame.

Last two bits (11) alerts that the next field is destination address.

preamble and SFD are added at the physical layer

and is not formally part of the frame

The Ethernet frame contains seven fields:


Ethernet (IEEE 802.3) MAC frame

DA : Destination address:

SA: Source Address:

Length/Type:

Define the upper-layer protocol using the MAC

frame. OR

define the number of bytes in the data filed.

Data: minumum: 46 and maximum : 1500 bytes

CRC: error detection information:CRC-32

12/6/2010

7


Frame length:

Minimum: 64 bytes (512 bits)

Maximum: 1518 bytes (12,144 bits)

Note


Minimum and maximum frame lengths

Minimum frame length restriction (64 bytes) is required for the correct

operation of CSMA/CD.

Min data length =64 - 18 (6+-6+2+4) = 46 bytes

If the upper- layer packet is less than 46 bytes, padding is added to

make up the difference.

Maximum length restriction; two historical reasons:

Memory was very expensive when Ethernet was designed.

Prevents one station from monopolizing the shared medium,

blocking other stations that have data to sent.

Max data length =1518- 18= 1500 bytes.

12/6/2010

8


Ethernet address in hexadecimal notation

Each station (PC or printer) has a network

interface card (NIC) which provides the station with

a 6-byte [48 bits] physical address (MAC adress)

It is written in hexadecimal notation, with a

colon between the bytes.


Unicast and multicast addresses

Source address is always a unicast address – the frames

comes from only one station.

Destination address can be:

unicast: defines only one recipient; one to one

multicast: a group of addresses; one to many

Broadcast: the recipients are all the stations on

the LAN

12/6/2010

9


The least significant bit of the first byte

defines the type of address.

If the bit is 0, the address is unicast;

otherwise, it is multicast.

The broadcast destination address is a

special case of the multicast address in which

all bits are 1s.

Note


Define the type of the following destination addresses:

a. 4A:30:10:21:10:1A b. 47:20:1B:2E:08:EE

c. FF:FF:FF:FF:FF:FF

SolutionTo find the type of the address, we need to look at the second

hexadecimal digit from the left. If it is even, the address is

unicast. If it is odd, the address is multicast. If all digits are F’s,

the address is broadcast. Therefore, we have the following:

a. This is a unicast address because A in binary is 1010.

b. This is a multicast address because 7 in binary is 0111.

c. This is a broadcast address because all digits are F’s.

Example 1

12/6/2010

10


Show how the address 47:20:1B:2E:08:EE is sent out on

line.

Solution

The address is sent left-to-right, byte by byte; for each

byte, it is sent right-to-left( LSB first), bit by bit, as

shown below:

Example 2

left-to-right :47→20→1B→2E→08→ EE 47 is 0100 0111 right -to-left 1110 0010


Slot Time

Slot time defined in bits:

It is the time required for a station to send 512 bits

(min frame size).

Slot Time = round trip time + time required to

send the jam sequence

It depends on the data rate, for traditional 10-Mbps

Ethernet it is 51.2 microseconds (512/10Mbps)

The choice of 512-bit Slot Time to allow the proper

function of CSMA/CD.

12/6/2010

11


Slot Time

If the sender sends a frame larger than the

minimum size ( 512 to 1518 bits)

If the station has sent out the first 512 bits and has not

heard a collision, it is guaranteed that the collision will

never occur during the transmission of the frame.

The reason is that all stations sensed the existence of the

signal and refrained from sending .

Collisions can only occur during the first half of the slot

time (slot time/2), and if it dose, it can be sensed by the

sender during the slot time.


Slot Time and Maximum network length

Slot time = 2 x Tp ( neglecting time required to send jam signal)

= 2 x Max Length/propagation speed

Max.Length = propagation speed x (SlotTime/2)

Let propagation speed = 2x108 m/s

Max.Length = (2 x 108) x (51.2 x10-6/2) = 5120 m

Consider the delay times in repeaters and interfaces, and

the time required to send the jam sequence.

Max.Length = 2500 m ( = 48 % of the theoretical)

12/6/2010

12


Categories of Standard Ethernet

The Standard Ethernet defines several physical layer implementation,

four of the most common:

Limitation of 10Base5 and 10Base2 is that communication

is half-duplex.

Why half deplux??


Encoding in a Standard Ethernet implementation

All standard implementations use digital signaling( baseband) at

10 Mbps.

At the sender, data are converted to a digital signal using the

Manchester scheme.

At the receiver, the received signal is interpreted as Manchester and

decoded into data.

Uses CSMA/CD with 1-persistent

12/6/2010

13


10Base5:Thick Ethernet (Thicknet)

Uses coaxial cable and Bus topology

With an external transceiver( transmitter/receiver) connected via a tap.

Transceiver is responsible for:

transmitting, receiving and

detecting collisions.

The length of each segment cannot exceed 500 m

If cable > 500 m ,degradation in the signal, using repeaters to connect

multiple “segments” of cable.

No two stations can be separated by more than 2500m( max length of the

bus) and 4 repeaters.


10Base2:Thin Ethernet (Cheapernet)

Uses Bus topology with thinner and more flexible cables.

Transceiver part of a NIC card

The implementation is more cost effective than 10Base5 because:

Thin coaxial cable is less expensive than the thick

tee connections are cheaper than taps

Installation is simpler because thin coaxial cable is very flexible.

The length of each segment under 200 (cannot exceed 185 m) due to the

high level of attenuation

Repeaters are used to connect multiple segments

12/6/2010

14


10 Base-T: Twisted-Pair Ethernet

Physical star topology

Stations are connected to a hub via two pairs of twisted

cable( one for sending and one for receiving|).

Any collisions happens in the hub

Compared to others, the hub replaces the coaxial cable as

far as a collision is concerned.

Max length = 100 m to minimize attenuation.


10Base-F: Fiber Ethernet

Uses star topology to connect stations to a hub

Stations is connected to the hub by using two

pairs of fiber-optic cables.

12/6/2010

15


Summary of Standard Ethernet

implementations

10Base 5 10Base2 10Base-T 10Base-F

Media Thick coaxial cable

Thin coaxial cable

Two UTP 2 Fiber

Maximum length

500 m 185 m 100 m 2000 m

Topology Bus Bus Star star

Data rate 10Mbps 10Mbps 10Mbps 10Mbps

Line coding Manchester Manchester Manchester Manchester


1313--3 3 CHANGES IN THE STANDARDCHANGES IN THE STANDARD

TheThe 1010--MbpsMbps StandardStandard EthernetEthernet hashas gonegone throughthrough

severalseveral changeschanges beforebefore movingmoving toto thethe higherhigher datadata

ratesrates.. TheseThese changeschanges actuallyactually openedopened thethe roadroad toto thethe

evolutionevolution ofof thethe EthernetEthernet toto becomebecome compatiblecompatible withwith

otherother highhigh--datadata--raterate LANsLANs..

Bridged Ethernet

Switched Ethernet

Full-Duplex Ethernet

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

12/6/2010

16


Bridged Ethernet

Have two effects on an Ethernet LAN:

1. They rise the bandwidth

2. They separate collision domains.

Without bridges, all the stations share the bandwidth of the

network.


Bridged Ethernet: Raising the Bandwidth

Bridges divide the network into two.

Each network is independent.

With bridges, 10 Mbps network is shared only by 6 [actually

7 as bridge acts as one station] stations.

12/6/2010

17


Using bridges: Collision domain becomes much smaller and the

probability of collision is reduced

Raising the B-W

Bridged Ethernet: Separate Collision domains


Switched Ethernet

A layer 2 switch is an N-port bridge with

additional sophistication that allows faster

handling of the packets.

12/6/2010

18


Full-duplex switched Ethernet

In full duplex switch there are two links, one for sending

and one for receiving, we don’t need CSMA/CD here|(

no collision).

Increases the capacity of each domain from 10 to

20 Mbps.


FAST ETHERNETFAST ETHERNET

FastFast EthernetEthernet waswas designeddesigned toto competecompete withwith LANLAN

protocolsprotocols suchsuch asas FDDIFDDI oror FiberFiber ChannelChannel.. IEEEIEEE

createdcreated FastFast EthernetEthernet underunder thethe namename 802802..33uu.. FastFast

EthernetEthernet isis backwardbackward--compatiblecompatible withwith StandardStandard

Ethernet,Ethernet, butbut itit cancan transmittransmit datadata 1010 timestimes fasterfaster atat

aa raterate ofof 100100 MbpsMbps..

12/6/2010

19


MAC Sublayer

Main consideration in the evolution of Ethernet from 10 to

100 Mbps was to keep the MAC sublayer untouched.

Drop bus topologies and keep only the star topology.

Why?? Star topology have two choices :

Half-duplex approach:

The stations are connected via a hub.

The access method is CSMA/CD

Full-duplex approach.(Fast )

The connection is made via a switch with buffers at

each port.

No need for CSMA/CD


Fast Ethernet topology

Topology:

Two stations: point-to-point

Three or more stations: star topology with a switch

at the center

12/6/2010

20


1. Upgrade the data rate to 100 Mbps

2. Make it compatible with Standard Ethernet

3. Keep the same frame format

4. Keep the same minimum and maximum frame length

Autonegotiation:

Allows two devices to negotiate the mode (half duplex

or full duplex) or data rate of operation.

Ex: Allow incompatible devices to connect to one another. A

device with a maximum capacity of 10 Mbps can

communicate with device with a 100 Mbps capacity (but

can work at a lower rate)

The goals of Fast Ethernet :The goals of Fast Ethernet :


Fast Ethernet implementations

Can be categorized as

Two wire: Category 5 UTP (100Base-TX)

Fiber-optic cable(100Base-FX)

four wire:Category 3 or higher UTP( 100Base-T4)

12/6/2010

21


Fast Ethernet implementations

Encoding:Manchester encoding needs a 200- Mbaud bandwidth for

a data rate of 100 Mbps.(unsuitable for a medium such as

Twisted Pair cable).

In Fast Ethernet , three different encoding schemes were

chosen depending on the implementation:

100 Base-TX

100Base-FX

100BAse-T4


100Base-TX

Uses two pairs of twisted-pair

cable (either category 5 UTP or

STP).

Provide data rate of 100 Mbps

MLT-3 scheme was selected

since it has good bandwidth

performance. (not self-

synchronization line coding)

So 4B/5B block coding is used

to provide bit synchronization.

12/6/2010

22


100Base-FX

Uses two pairs of fiber-optic

cables.

Uses NRZ-I encoding

scheme ( bit synchronization

problem.)

To overcome this problem,

4B/5B block coding is used.


100Base-T4

Uses four pairs of

category 3 or higher

UTP.(not cost efficient

compared to Category 5)

Transmit 100 Mbps.

Uses 8B/6T encoding

12/6/2010

23


Summary of Fast Ethernet implementations

100Base -TX 10Base-FX 10Base-T4

Media Cat 5 UTP or STP

Fiber Cat 3 UTP or higher

Number of wires 2 2 4

Maximum length 100 m 400 m: half duplex2000m: full duplex

100 m

Topology Star Star Star

Data rate 100 Mbps 100 Mbps 100 Mbps

Block encoding 4B/5B 4B/5B 8B/6T

Line encoding MLT-3 NRZ-I


1313--5 5 GIGABIT ETHERNETGIGABIT ETHERNET

TheThe needneed forfor anan eveneven higherhigher datadata raterate resultedresulted inin thethe

designdesign ofof thethe GigabitGigabit EthernetEthernet protocolprotocol ((10001000 Mbps)Mbps)..

TheThe IEEEIEEE committeecommittee callscalls thethe standardstandard 802802..33zz..

Documents

Wired LANs: Ethernet - Erode Sengunthar Engineering … LANs: Ethernet Chapter 13 Dr. Gihan NAGUIB ... If the upper- layer packet is less than 46 bytes, ... network. Dr. Gihan NAGUIB