2.- Ethernet system

Local Area Networks/School of Engineering in Computer Science/2009-2010

http

://w

ww.r

edes

.upv

.es/

rali

r/en

/2.- Ethernet system

Other technologies The physical layer: a reminder general aspects

MAC Full-duplex auto-negotiation

Ethernet Fast Ethernet Gigabit Ethernet

thanks to:Igor Radovanović,

[email protected] Technische Universiteit Eindhoven


http

://w

ww.r

edes

.upv

.es/

rali

r/en

/Other LAN technologies

Token RingToken BusFDDI

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce

http://www.redes.upv.es/ralir/en/

What is a Token Ring LAN?

A token ring network (IEEE 802.5) is a LAN technology based on a token-passing protocol for media access control. (No collisions)

Data frames are transmitted from node to node, in either a clockwise or counter-clockwise direction, over a point-to-point link.

A Token Ring LAN is implemented either: Logical ring using a physical ring topology Logical ring using a physical star topology (uses Token Ring Hubs

that incorporate logical rings)

3

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Token-Passing Access Methodology

4

Grabs 24 bit token and sets T-bit from 1 to 0 (free to busy) Puts data frame on network. Sets destination:Station D is in transmit mode.

Receives frame and reads its destination address Sees destination is Station D and copies the received frame to the output link is in listen mode.

is powered off so network traffic flows through unprocessed. is in bypass mode.

Receives frame and identifies itself as the destination address Moves data from frame into memory Changes frame status flags to confirm delivery

Receives originally sent frame Examines frame status flags to find: A=1: address recognized C=1: frame copied successfully Does NOT retransmit frame Sets T-bit in token to 1 (free) Releases token

FR

AM

EA

CD

ATA

SD

DA

00

AD

10010101T0

FRAMEA C DATA SD DA0 0 A D10010101

T0

FR

AM

EA

CD

ATA

SD

DA

00

AD

1001

0101

T 0

FRAMEA C DATA SD DA1 1 A D10010101

T0Station A:

Starts here

Station D:

Station C:Station B:

TACDATASDDA

busy statusaddress recognizedcopied successfullytransmitted datasource addressdestination address

ABBREVIATED FRAME LEGEND

Ends here when Station A:

GOLDMAN & RAWLES: ADC3e FIG. 05-08

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Token Ring Topologies

5

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Terminology

Token Ring hubs are called Multistation Access Units (MAUs).

Nodes connected to a Token Ring network are called lobes;

The distance between MAUs is called the main ring length.

The distance between an MAU and its lobes is called lobe length.

IEEE 802.5 defined the specifications of Token Ring LANs, which is based in part on IBM’s set of token ring specifications. (IBM is the primary vendor of this technology)

6

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Illustrative Example of MAUs

7

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Token Ring vs. Token Bus (1)

A Token Bus is a logical ring on a physical bus topology. It uses a token passing protocol (defined in IEEE 802.4).

The network is physically a bus topology and logically arranged as a ring with respect to passing the token from lobe to lobe.

Token Bus is based on broadcasting (similar to Ethernet/802.3), even thought token passing is point-to-point.

The Medium Access Control is similar to Token Ring (802.5).

At the physical layer, IEEE 802.4 uses coaxial cable or fiber-optic cable.

8

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Token Ring vs. Token Bus (2)

Token Bus priority scheduling supports up to four priority modes (0 (lowest), 2, 4, and 6 (highest)).

Token Bus is not a popular LAN protocol, primarily used for control of industrial and factory automation processes.

9

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Advantages and Disadvantages

Advantages: its ability to run on many different media types its efficient use of bandwidth its stable behavior during high load times its deterministic nature its priority scheme, which enables nodes with high-priority data to

reserve the network Disadvantages:

need special recovery procedures when network fails difficulty in configuring new hosts to an established LAN in priority scheduling, the susceptibility of low priority nodes to

increased delays in accessing the network.

10

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


FDDI: General aspects

Fiber Distributed Data Interface MAC: Token Passing Logical topology: Sequential Physical topology: Dual counter-rotating rings. Uses cable types: Fiber optic cable Data rate/speed = 100 Mbps

Each cable length = 2 Km Max # of cables = 100; Max # of repeaters = 100 Max total Network length = 200 Km Max # of nodes/segment = 10 Max # of nodes/network = 1000

11

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


FDDI: Architecture basics

Primary ring---is the outer ring. Secondary ring---is the inner ring. Uses 2-types of Computers: DAS and SAS SAS = Single attachment station

Connected to the outer/primary ring DAS = Dual attachment station

Connected to both the primary/secondary rings. If SAS/Primary ring fails, DAS takes over. Each of these DASs/SASs must use FDDI-NIC.

12

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


FDDI-Topology

13


http

://w

ww.r

edes

.upv

.es/

rali

r/en

/2.- Ethernet

The physical layer: a remindergeneral aspects

MAC Full-duplex auto-negotiation

EthernetFast EthernetGigabit Ethernet

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Physical layer reminder: Terminology

Guided (wired) vs. Unguided (wireless) examples of guided transmission media:

Twisted Pair Coaxial cable Optical fiber

point-to-point vs. multipoint

simplex, half-duplex and full-duplex links15

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Physical layer reminder: Twisted Pair

pair short connections (< 50m) data rates < 19.2 kbps Problems:

crosstalk noise

twisted pair often “bundled” into cables

and usually installed in buildings during constructions

better than pair but still susceptible to interference and noise

Pros and Cons: Cheap Easy to work with Limited data rate (100MHz) Limited bandwidth (1MHz) Limited distance

repeater every 2km or 3km

16

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Physical layer reminder: Unshielded and Shielded Twisted Pair

Unshielded Twisted Pair (UTP) Cheapest Easiest to install Suffers from external EM interference Categories:

Cat 3 up to 16MHz Voice grade found in most offices Twist length of 7.5 cm to 10 cm

Cat 4 up to 20 MHz

Cat 5 up to 100MHz Commonly pre-installed in new office buildings Twist length 0.6 cm to 0.85 cm

Shielded Twisted Pair (STP) Metal braid or sheathing that reduces interference More expensive Harder to handle (thick, heavy)

17

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Physical layer reminder: Coaxial Cable

Most versatile medium two types: 50-ohm and 75-ohm

Long distance telephone transmission Can carry 10,000 voice calls

simultaneously Being replaced by fiber optic

Short distance computer systems links

Local area networks

Analog Amplifiers every few km Closer if higher frequency up to 500MHz

Digital Repeater every 1km Closer for higher data rates up to 350 MHz (500 Mbps)

18

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Physical layer reminder: Optical Fiber

Greater capacity Data rates of hundreds of

Gbps Smaller size & weight Lower attenuation Electromagnetic isolation Greater repeater spacing

10s of km at least problems with junctions

Applications: Long-haul trunks Metropolitan trunks Rural exchange trunks Subscriber loops LANs

19

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Physical layer reminder: Optical Fiber - Transmission Characteristics

Act as wave guide for 1014 to 1015 Hz Portions of

infrared and visible spectrum

Light Emitting Diode (LED) Cheaper Wider operating

temp range Last longer

Injection Laser Diode (ILD) More efficient Greater data rate

Wavelength Division Multiplexing: WDM and DWDM

20

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Physical layer reminder: Encoding and modulation techniques

21

coder

x(t) [digital]

g(t) [digital or analog]

decoder

g(t)

modulator

s(t) [analog]

m(t) [digital or analog]

demodulator

m(t)

fc

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Physical layer reminder: Nonreturn to Zero-Level

NRZ-L: two different voltages for 0 and 1 bits

Voltage constant during bit interval no transition, i.e., no return to

zero voltage Example: absence of voltage

for zero, constant positive voltage for one

More often, negative voltage for one value and positive for the other

NRZ-I: Nonreturn to zero inverted on ones

Constant voltage pulse for duration of bit

Data encoded as presence or absence of signal transition at beginning of bit time

Transition (low to high or high to low) denotes a binary 1; no transition denotes binary 0

22

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Physical layer reminder: Self-clocking line encoding

23

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Physical layer reminder: Multilevel Binary

24

mBnL codes: a sequence of m bits is represented by n pulses, each of L levels (n<m and L>2)

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Ethernet: some historical notes...

1972: Robert Metcalfe designs the first prototype at the Xerox PARC

1980: the first standard appears, by DEC-Intel-Xerox 1985: IEEE802.3 standard is released. The complete

name is: “Carrier Sense Multiple Access with Collision Detection (CSMA/CD) Access Method and Physical Layer Specifications”

IEEE supplements 802.3d-1987: FOIRL fiber link 802.3i-1990: 10BASE-T twisted pair 802.3z-1998: Gigabit Ethernet 802.3ab-1999: 1000BASE-T Gigabit Ethernet over twisted pair 802.3ae-2002: 10Gb/s Ethernet

25

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


The Ethernet “family tree”

10Mbps 10Base2 10Base5 10Base-T FOIRL 10Base-F

10Base-FB 10Base-FP 10Base-FL

100 Mbps (100Base-T) 100Base-X

100Base-TX 100Base-FX

100Base-T4 1000 Mbps

1000Base-X 1000Base-SX 1000Base-LX 1000Base-CX

1000Base-T

26

2 = thin coaxial; 5 = thick coaxial; T = twisted pair; F = fiber optics;

S = short wavelength; L = long wavelength; C = short copper cable;

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Devices

Hubs

Switches

Adaptors

27

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Connection media

Different connection types are used by each physical layer implementation.

The registered jack (RJ-45) connector and jack are the most common.

28

In some cases the type of connector on a NIC does not match the media that it needs to connect to. The AUI connector allows different media to connect when used with the appropriate transceiver.

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Connection media

EIA/TIA specifies an RJ-45 connector for UTP cable. The letters RJ stand for registered jack, and the number

45 refers to a specific wiring sequence. The RJ-45 connector is the male component, crimped on

the end of the cable.

29

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Connection media

The jack is the female component in a network device, wall outlet, or patch panel.

30

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Cable Standards

For electricity to run between the connector and the jack, the order of the wires must follow EIA/TIA-T568-A or T568-B standards.

Identify the correct EIA/TIA category of cable to use for a connecting device by determining what standard is being used by the jack on the network device.

31

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Straight-through Cable

The 10BASE-T and 100BASE-TX Ethernet standards use one wire pair for transmission in each direction. This requires that the transmit pair of each device be connected to the receive pair of the device on the other end.

If the two RJ-45 connectors of a cable are held side by side in the same orientation, the colored wires will be seen in each.

If the order of the colored wires is the same at each end, then the cable is straight-through.

32

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Straight-through Cable

Use straight-through cables for the following cabling: Switch to Host (PC or server) Hub to Host (PC or server) Switch to router

33

Rule of thumb: Connect devices of the same OSI Layer with CROSSOVER cable.Connect devices of different OSI Layers with STRAIGHT cables.

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Crossover cables

One terminal device may be connected directly to another without the use of a switch or hub, but in that case the crossover must be done externally in the cable. Since 10BASE-T and 100BASE-TX use pairs 2 and 3, these two pairs must be swapped in the cable. This is a crossover cable. When a terminal device is connected to a switch or hub, this crossover is done

internally in the switch or hub. Automatic crossover

Automatic MDI/MDI-X Configuration is specified as an optional feature in the 1000BASE-T standard meaning that straight-through cables will usually work between Gigabit capable interfaces. This feature eliminates the need for crossover cables, obsoletes the uplink/normal ports and manual selector switches found on many older hubs and switches, greatly reducing installation errors.

Note that although Automatic MDI/MDI-X is generally implemented, a crossover cable would still be required in the occasional situation that neither of the connected devices has the feature implemented and enabled.

Even for legacy 10/100 devices, many NICs, switches and hubs automatically apply an internal crossover when necessary. This feature may also be referred to by various vendor-specific terms including: Auto uplink and trade, Universal Cable Recognition and Auto Sensing.

34

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Crossover Cable

With crossover, the RJ-45 connectors on both ends show that some of the wires on one side of the cable are crossed to a different pin on the other side of the cable.

Pins 1 and 2 on one connector connect respectively to pins 3 and 6 on the other.

35

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Crossover Cable

Use crossover cables typically between like devices, for example: Switch to switch Hub to hub Router to router PC to PC

36

Rule of thumb: Connect devices of the same OSI Layer with CROSSOVER cable.Connect devices of different OSI Layers with STRAIGHT cables.

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Frame format in traditional Ethernet MAC frame

Preamble 7 bytes of alternating 0s and 1s – receiver sync Not necessary in high-speed Ethernet systems (FastEth, GigE)

Start of Frame Delimiter 10101011 – unique sequence -> last chance to synchronize

Source Address, Destination Address 48 bit unique address

Length/Type Value up to 1518: length; value larger than 1536 – type of PDU encaps

Data Frame Check Sequence (CRC); Preamble and SFD excluded

37

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Min & Max Ethernet frame length

Minimum length - used for CSMA/CD Every end station senses the frame within the correct time limits Historical requirement derived for bus topology with a coax cable

and a network with a 2 byte field Maximum length – to assure fair access

A station should not occupy the medium too long

38

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


48-Bit Universal LAN MAC Addresses

Described in IEEE Std802-1990, pg. 25 48 bits burned into ROM on adaptor

Unicast address: destined for individual receiver Broadcast address: all 1’s in address- for every node Multicast address: leading 1 in address– subset of nodes

39

Organizationally Unique Identifier as assigned by the IEEE

Octet: 0 1 2 3 4 5 0011 0101 0111 1011 0001 0010 0000 0000 0000 0000 0000 0001

First bit transmitted on the LAN medium. (Also the I/G Address Bit: 0=Individual; 1=Group)Second bit, the Universally or Locally Administered, U/L Address Bit, indicates whether the address has been assigned by a local (=1) or universal administrator (=0).

The hexadecimal representation is: AC:DE:48:00:00:80

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Random Access MAC protocols CSMA/CD (Collision Detection)

CSMA/CD: carrier sensing, deferral as in CSMA collisions detected within short time colliding transmissions aborted, reducing channel wastage persistent or non-persistent retransmission

collision detection: easy in wired LANs: measure signal strengths, compare

transmitted, received signals difficult in wireless LANs: receiver shut off while transmitting

human analogy: the polite conversationalist

40

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Random Access MAC protocols CSMA/CD collision detection

41

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


The CSMA with Collision Detection protocol

42

while there are frames to be sentwhile (channel is busy)loop;/* the channel is available (idle) */wait IFG<< sending >>if a collision is detected then

send jam signalabort the transmission

wait a random amount of timewhend

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Binary Exponential Backoff

43

-> ith collision <-

if i 10 thenchoose a number n between 0 and (2i-1)wait n time-slotstry sending again

else if i 16 thenchoose a number n between 0 and (210-1)wait n time-slots

try sending again

else if i > 16 thenerror

fi

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


CSMA/CD: a few details

44

time

A

B

C

datainterframe gap (IFG): 96 bits (9.6msec at 10Mbps)collision enforcement jam signal: 32 bitscollisionbackOff time: multiples of the slot time: 512 bits

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Collisions and minimum frame size

45

t=0 t = t-e

A B A B

t = t

A B

t = 2t

A B

1km coax cable, Tprop = d/Vprop ~ 5msec; 2,5 km with 4 repeaters, measured 2t ~ 51.2msecs = vt = 10Mbps·51.2msec = 512 bits=64bytes

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Media System Timing

Depends on: physical layer roud-trip propagation time collision enforcement procedure

512 bit time for 10 and 100 Mbps sytems networks diameter in 10 Mbps systems: approx. 2800meters networks diameter in 100 Mbps systems: approx. 205meters

In half-duplex Gigabit ethernet networks diameter would be approx. 20 meters The solution: carrier extension; it allows not to modify the frame

structure

46

i.e., 4096 bits

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Frame bursting

Carrier extension is very inefficient for the short packets for example for 64 byte packet we have 448 bytes of padding

low throughput Solution: introduce Frame bursting

The maximum burst size is 65536 bit (8192 bytes) plus the final frame transmission

47

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Definition of a collision domain

“defined as all the Ethernet segments between a pair of bridges or other layer 2 devices”

“a single CSMA/CD network in which there will be a collision if two computers attached to the system both transmit at the same time”

48

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Segmentation

Segmentation is the process of splitting a single collision domain into two or more collision domains. Layer 2 bridges or switches can be used to segment a logical bus topology and create separate collision domains. However the entire bus topology still represents a single broadcast domain because, bridges and switches forward broadcast packets.

49

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


An example of a collision domain

A collision domain is also known as a bandwidth domain

50

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Bridged Ethernet: creating separate collision domains

Bridges: rise the bandwidth and separate collision domains

51

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Switched Ethernet: micro-segmentation

Each end station uses a separate path to the port in the switch (no sharing medium)

Used to add bandwidth without replacing NICs

52

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Full-duplex Ethernet

“Ethernet is CSMA/CD” Dedicated media (structured wiring) and micro-

segmentation (provided by switches) have evolved to enable full-duplex capability in Ethernet

Disable carrier-sense, collision detect! Eliminate the link length restrictions Increase the aggregate channel capacity

Gigabit Ethernet basically works in full-duplex mode

53

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Limitations of Full-duplex Operation

Requirements for full duplex as stated in the 802.3x standard:1. The underlying physical channel must be capable of

supporting simultaneous, bi-directional communications without interference

2. Exactly two devices on the LAN segment.3. The interfaces in both devices must be capable of

and configured to use full-duplex mode.

If all of these conditions are met, then full-duplex mode not only can be used, it should be used.

54

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Ethernet Flow Control

Need for flow control Loss of a frame due to buffer unavailability has the same effect as

a frame lost due to a bit error Effect of frame loss can be devastating on throughput if flow

control is implemented by a higher layer protocol Backpressure in half duplex networks will not work; a full duplex

Ethernet requires an explicit flow control

MAC Control client (such as a switch desiring to prevent buffer overflow) uses the added capabilities of the MAC control sub layer.

It can request the MAC at the other end of a full-duplex link to cease further data transmissions, thereby preventing the impending overflow (PAUSE function)5

5

MAC Control

Data LinkPhysical

Network

MAC Control Client

MAC – Media Access Control(higher layers)

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


PAUSE Function

Implement flow control on a single full-duplex Ethernet link

Simple “stop-start” form of flow-control A device (end station or switch) wishing to temporarily inhibit

incoming data sends a PAUSE frame (along with the time to wait) When a device receives a PAUSE frame, it stops sending data

frames for the period specified Flow control policy

56

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


PAUSE Packet

The only packet defined by the MAC control sublayer A station starts a timer set to the specified pause time

after it receives a PAUSE packet and resumes sending data frames when the timer expires

A station can still send PAUSE packet after it receives a PAUSE frame to ask the other device to stop sending data

A device may send several PAUSE packets. The more recent PAUSE packet can cancel the effect of the previous one

57

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Auto negotiation

The auto-negotiation algorithm (known as NWay) allows two devices at either end of a 10 Mbps, 100 Mbps, or 1000 Mbps link to advertise and negotiate the link operational mode—such as the speed of the link and the duplex configuration of half or full duplex—to the highest common denominator.

Auto-negotiation addresses the following goals: Provide easy, plug-and-play upgrades from 10 Mbps, 100 Mbps,

and 1000 Mbps as the network infrastructure is upgraded Prevent network disruptions when connecting mixed technologies

such as 10BaseT, 100BaseTX, and 1000BaseT Accommodate future PHY (transceiver) solutions Support backward compatibility with 10BaseT

The negotiation covers only the link (between a station and a hub or between two hubs), not the whole network

Negotiation can only occur during link initialization Negotiation uses a separate frame format and signaling

system Each device advertises its capabilities to the other

device and the decision is based on common capabilities58

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Auto negotiation

The Auto-Negotiation specification includes reception, arbitration, and transmission of normal link pulses (NLPs). It also defines a receive LIT function for backward compatibility with 10BaseT devices.

All of these functions are implemented as part of the physical layer transceiver. The exchange of link information occurs between the PHY and the Medium Dependent Interface (MDI) or RJ-45 connector.

All auto-negotiation functions are implemented as part of the transceiver integrated circuit, which is part of a network interface card or integrated on the motherboard of a computer.

59

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Auto negotiation: link test 10BaseT link test pulses

The 10BaseT standard includes a link test mechanism to ensure network integrity. In the absence of network traffic, a 100 nanosecond (ns) heartbeat unipolar (positive only) pulse is sent every 16 milliseconds (ms) within a range of +/- 8 ms. The link test pulse is sent by the transmitters of all 10BaseT media attachment units (MAUs) between the data terminal equipment (DTE) and the repeater.

A link fail condition is entered if the receiver does not receive a packet or a link test pulse within 50-150 ms. The link fail condition disables the data transmit, data receive, and loopback functions. The link test pulses continue to be transmitted and received during the link failure. The link is reestablished when two consecutive link test pulses or a single data packet have been received.

100BaseT/1000BaseT fast link pulses The link information is encoded in

a special pulse train known as the fast link pulse (FLP) burst. The FLP builds on the LIT pulse used by 10BaseT devices as a heartbeat pulse to the link partner at the opposite end of the link. The LIT was redefined as the normal link pulse (NLP).

The NLP is the 10BaseT link integrity test pulse, and the FLP is a group of NLPs. Each pulse is 100 ns in width.

Auto-negotiation replaces the single 10BaseT link pulse with the FLP burst. Auto-negotiation stops the transmission of FLP bursts once the link configuration is established.

The FLP burst looks the same as a single link test pulse from the perspective of 10BaseT devices. Consequently, a device that uses NWay must recognize the NLP sequence from a 10BaseT link partner, cease transmission of FLP bursts, and enable the 10BaseT physical medium attachment (PMA).

60

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


FLP bursts FLP bursts

Each FLP burst consists of 33 pulse positions that provide clock and data information. The 17 odd-numbered pulses are designated as clock pulses, while the 16 even-numbered pulse positions represent data information. A logic one is represented by the presence of a pulse, while the absence of a pulse is represented by a logic zero.

FLP burst encoding The data pulses in the FLP burst

encode a 16-bit link code word (LCW). A device capable of auto-negotiation transmits and receives the FLP. The receiver must identify three identical LCWs before the information is authenticated and used in the arbitration process. The devices decode the base LCW and select capabilities of the highest common denominator supported by both devices. Once the LCWs are properly received, each device transmits a FLP burst with an acknowledge bit. At this point, both devices enable the mode that is the highest common mode negotiated.

61

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Three generations of Ethernet

62

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Ethernet- Physical layer

Medium Attachment Unit Medium dependant

MDI External: tap or a tee connector Internal: jack

63

If PLS to be connected to a different MAU PLS the same

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Ethernet 10Base5

64

(MAU) 10Base5 transceiver

MDI

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Ethernet 10Base2

65

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Physical layer implementation

66

sTYPE-t(l)

s – speed in MbpsTYPE – broadband or baseband signaling l – cable distance in multiple of 100 mt – media type used

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


10BASE-T

uses a physical star topology end stations connected to a hub using external or

internal transceiver maximum distance 100 m using UTP cable maximum number of hubs is 4 (total span 2500 m)

67

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


10BaseT: the RJ45 jack

68

“cross-over” connection

Pin#1: TX+Pin#2: TX-Pin#3: RX+Pin#6: RX-

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


10BaseT: twisted pair

69

HUB HUB

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce



70

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Fast Ethernet

Keep MAC layer untouched Increase the speed 10 times while keeping the

compatibility with the Ethernet Quick deployment Autonegotiation

negotiate the mode (full- or half-duplex) or data rate of operation (10 Mbps, 100 Mbps)

Different encoding technique to transfer the high data rate signal

Physical layer Reconciliation: pass data in 4-bit format (nibble) to the MII

interface MII: operates at both 10 Mbps and 100 Mbps; features a parallel

data path (4 bits) between PHY and reconciliation sublayer; defines 5 types of signals: transmit, receive, control (carrier sense & collision detect), management, and power; not needed for internal PHY (see page 242)

PHY: transceiver, also responsible for encoding/decoding MDI: medium dependent interface to connect to the medium7

1

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Fast Ethernet implementations

72

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


100BASE-TX

uses two pairs of twisted-pair cable (Cat5 UTP or STP) in the physical star topology (up to 500 m)

73

encoding

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


MLT-3

Uses 3 levels of signals (+1,0,-1) Transition at the beginning of bit 1 Used to decrease transmission frequency

Q: Why?

74

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Physical layer reminder: 4B5B coding

Used by 100Base-TX Based on NRZI at 125 Mbaud/s. 4 bits are represented with 5 bits symbols using a table. The remaining sequences are used for signaling

75

1 1 1 0 1 00 1+

_

0

+

_

0

8 ns. 125Mbaud/s

1 1 10 1 1 1 00 1

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


100BASE-FX

Two pairs of fiber-optic cables in a physical star topology

Can support larger distances (up to 2km); unrepeated 412 m single-mode fiber

Long-wavelength lasers in NICs Problem: not compatible with previous fiber standards

76

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


100BASE-FX - encoding

77

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce



78

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Gigabit Ethernet

Once more 10 times faster Ethernet wanted to keep MAC layer unaltered 802.3z (z indicates the end of the road) Design goals:

backwards compatibility offer unacknowledged datagram services same 48 bit address scheme

All configurations are point-to-point Defined for both half- & full-duplex Half-duplex supports CSMA/CD All implementations today are full-duplex

79

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Gigabit Ethernet Physical Layer

Reconciliation: pass data in 8-bit format to the PHY sublayer via a GMII interface

GMII: operates at 1000 Mbps; with chips supporting both 10 Mbps and 100 Mbps; features a parallel data path (8 bits) between PHY and reconciliation sublayer; a logical, rather than a physical interface

PHY: transceiver, also responsible for encoding/ decoding MDI: medium dependent

interface to connect to the medium

80

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


Gigabit Ethernet implementations

81

based on optical fiber

short wavelengths 260 m -> MM fiber 550 m -> SM fiber

long wavelengths 550 m -> MM fiber 3 km -> SM fiber

25 m

based on STP; never implemented

Cat5 UTP

100 m

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


1000BaseSX & LX

The NRZ signals are then pulsed into the fiber using either short-wavelength or long-wavelength light sources. The short-wavelength uses an 850 nm laser or LED source in

multimode optical fiber (1000BASE-SX). It is the lower-cost of the options but has shorter distances.

The long-wavelength 1310 nm laser source uses either single-mode or multimode optical fiber (1000BASE-LX). Laser sources used with single-mode fiber can achieve distances of up to 5000 meters.

Because of the length of time to completely turn the LED or laser on and off each time, the light is pulsed using low and high power. A logic zero is represented by low power, and a logic one by high power.

The Media Access Control method treats the link as point-to-point. Since separate fibers are used for transmitting (Tx) and receiving (Rx) the connection is inherently full duplex.

Gigabit Ethernet permits only a single repeater between two stations.

82

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


1000 Mbps on Cat5e? How?

Because Cat 5e cable can reliably carry up to 125 Mbps of traffic, getting 1000 Mbps (Gigabit) of bandwidth was a design challenge.

The first step to accomplish 1000BASE-T is to use all four pairs of wires instead of the traditional two pairs of wires used by 10BASE-T and 100BASE-TX. This is done using complex circuitry to allow full duplex

transmissions on the same wire pair. This provides 250 Mbps per pair.

With all four-wire pairs, this provides the desired 1000 Mbps. Since the information travels simultaneously across the

four paths, the circuitry has to divide frames at the transmitter and reassemble them at the receiver.

83

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


1000Base-T

84

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


10-Gigabit Ethernet - 2002

IEEE 802.3ae was adapted to include 10 Gbps full-duplex transmission over fiber optic cable.

This 10-Gigabit Ethernet (10GbE) is evolving for not only LANs, but also MANs, and WANs.

With the frame format and other Ethernet Layer 2 specifications compatible with previous standards, 10GbE can provide increased bandwidth needs that are interoperable with existing network infrastructure.

Ethernet is traditionally thought of as a LAN technology, but 10GbE physical layer standards allow both an extension in distance to 40 km over single-mode fiber and compatibility with synchronous optical network (SONET) and synchronous digital hierarchy (SDH) networks.

Compatibility with SONET/SDH networks operating up to OC-192 speeds (9.584640 Gbps) make 10GbE a viable WAN technology. 10GbE may also compete with ATM for certain applications.8

5

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


10-Gigabit Ethernet Architecture

As with the development of Gigabit Ethernet, the increase in speed comes with extra requirements. The shorter bit time duration because of increased speed requires

special considerations. Because of the short duration of the 10 GbE data bit, it is often difficult to separate a data bit from noise.

10 GbE data transmissions rely on exact bit timing to separate the data from the effects of noise on the physical layer. This is the purpose of synchronization.

By using codes to represent the user data, transmission is made more efficient. The encoded data provides synchronization, efficient usage of bandwidth, and improved Signal-to-Noise Ratio characteristics.

Complex serial bit streams are used for all versions of 10GbE except for 10GBASE-LX4, which uses Wide Wavelength Division Multiplex (WWDM) to multiplex four bit simultaneous bit streams as four wavelengths of light launched into the fiber at one time.

86

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


10-Gigabit Ethernet Variety

As is typical for new technologies, a variety of implementations are being considered, including:

10GBASE-SR Intended for short distances over already-installed multimode fiber

supports a range between 26 m to 82 m 10GBASE-LX4

Uses wavelength division multiplexing (WDM) supports 240 m to 300 m over already-installed multimode fiber and 10 km over single-mode fiber

10GBASE-LR and 10GBASE-ER Support 10 km and 40 km over single-mode fiber

10GBASE-SW, 10GBASE-LW, and 10GBASE-EW Known collectively as 10GBASE-W are intended to work with OC-

192 synchronous transport module (STM) SONET/SDH WAN equipment.

87

Loca

l Are

a Ne

twor

ks (R

ALIR

) /Sc

hool

of E

ngin

eerin

g in

Com

pute

r Scie

nce


The Future of Ethernet

Ethernet has gone through an evolution from Legacy → Fast → Gigabit → MultiGigabit technologies.

While other LAN technologies are still in place (legacy installations), Ethernet dominates new LAN installations.

While 1-Gigabit Ethernet is now widely available and 10-Gigabit products becoming more available, the IEEE and the 10-Gigabit Ethernet Alliance are working on 40, 100, or even 160 Gbps standards. The problem of collisions with physical bus topologies of 10BASE5

and 10BASE2 and 10BASE-T and 100BASE-TX hubs is no longer common. Using UTP and optical fiber with separate Tx and Rx paths, and the decreasing costs of switches make single shared media, half-duplex media connections much less important.

The future of networking media is three-fold: Copper (up to 1000 Mbps, perhaps more) Wireless (approaching 100 Mbps, perhaps more) Optical fiber (currently at 10,000 Mbps and soon to be more)

88

Documents

2.- Ethernet system