Chapter 13

Wired LANs: Ethernet

McGraw-Hill Copyright © The McGraw-Hill Companies, Inc. Permission required for reproduction or display.
 13-1 IEEE STANDARDS

Ethernet: It is a LAN protocol that is used in Bus and Star topologies

and implements CSMA/CD as the medium access method

Original (traditional) Ethernet developed in 1980 by

three companies: Digital, Intel, Xerox (DIX).

In 1985, the Computer Society of the IEEE started a

project, called Project 802, to set standards to enable
intercommunication among equipment from a variety of
manufacturers.
 Current version is called IEEE Ethernet
McGraw-Hill
Ethernet

 Ethernet Frame format

FCS

FCS

Frame formats. (a) DIX Ethernet , (b) IEEE 802.3.

McGraw-Hill
 Figure 13.4 802.3 MAC frame

McGraw-Hill
 IEEE Ethernet
 In IEEE 802.3 Ethernet Data link layer is split into two sublayers:
 Bottom part: MAC
 The frame is called IEEE 802.3
 Handles framing, MAC addressing, Medium Access control
 Specific implementation for each LAN protocol
 Defines CSMA/CD as the access method for Ethernet LANs and Token passing
method for Token Ring.
 Implemented in hardware
 Top part: LLC (Logical Link Control)
 The subframe is called IEEE 802.2
 Provides error and flow control if needed
 It makes the MAC sublayer transparent
 Allows interconnectivity between different LANs data link layers
 Used to multiplex multiple network layer protocols in the data link layer
frame
 Implemented in software

McGraw-Hill
 Figure 13.1 IEEE standard for LANs

McGraw-Hill
Ethernet Provides Unreliable, connectionless Service
 Ethernet data link layer protocol provides
connectionless service to the network layer
 No handshaking between sending and receiving

adapter.
 Ethernet protocol provides Unreliable service to the
network layer :
 Receiving adapter doesn’t send ACK or NAK to

sending adapter
 This means stream of datagrams passed to network

layer can have gaps (missing data)

 Gaps will be filled if application is using reliable transport
layer protocol
 Otherwise, application will see the gaps

McGraw-Hill
 Ethernet Frame
 Preamble:
 8 bytes with pattern 10101010 used to synchronize receiver, sender clock rates.
 In IEEE 802.3, eighth byte is start of frame (10101011)
 Addresses: 6 bytes (explained latter)
 Type (DIX)
 Indicates the type of the Network layer protocol being carried in the payload
(data) field, mostly IP but others may be supported such as IP (0800), Novell IPX
(8137) and AppleTalk (809B), ARP (0806) )
 Allow multiple network layer protocols to be supported on a single machine
(multiplexing)
 Its value starts at 0600h (=1536 in decimal)
 Length (IEEE 802.3): number of bytes in the data field.
 Maximum 1500 bytes (= 05DCh)
 CRC: checked at receiver, if error is detected, the frame is discarded
 CRC-32
 Data: carries data encapsulated from the upper-layer protocols
 Pad: Zeros are added to the data field to make the minimum data length = 46 bytes

McGraw-Hill
 Ethernet address
 Six bytes = 48 bits
 Flat address not hierarchical
 Burned into the NIC ROM
 First three bytes from left specify the vendor. Cisco
00-00-0C, 3Com 02-60-8C and the last 24 bit should be
created uniquely by the company
 Destination Address can be:
 Unicast: second digit from left is even (one
recipient)
 Multicast: Second digit from left is odd (group of
stations to receive the frame – conferencing
applications)
 Broadcast (ALL ones) (all stations receive the
frame)
 Source address is always Unicast
McGraw-Hill
McGraw-Hill
 Note

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.

McGraw-Hill
 Note

The broadcast destination address is a

special case of the multicast address in
which all bits are 1s.

McGraw-Hill
 Figure 13.7 Unicast and multicast addresses

McGraw-Hill
 Example 13.2

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, bit by bit, as shown below:

McGraw-Hill
 Example 13.1

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
Solution
To 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.
McGraw-Hill
 Figure 13.5 Minimum and maximum lengths

McGraw-Hill
 Note

Frame length:
Minimum: 64 bytes (512 bits)
Maximum: 1518 bytes (12,144 bits)

McGraw-Hill
 Figure 13.3 Ethernet evolution through four generations

McGraw-Hill
 Categories of traditional Ethernet

•<data rate><Signaling method><Max segment length or cable type>

McGraw-Hill
 IEEE 802.3 Cable Types
Name Cable Max. Max Cable Nodes
Toplogy
Segment /segment
Length
10Base5 thick coax 500 meters 100
Bus

10Base2 thin coax 185 meters 30

Bus

10BaseT twisted pair 100 meters 1 Star

10BaseF Fiber Optic 2Km 1

Star

McGraw-Hill
 Figure 13.10 10Base5 implementation

McGraw-Hill
 Connection of stations to the medium using 10Base2

McGraw-Hill
 10BaseT
• Uses twisted pair Cat3 cable
 Star-wire topology

• A hub functions as a repeater with additional functions

• Fewer cable problems, easier to troubleshoot than coax
• Cable length at most 100 meters

McGraw-Hill
 Figure 13.12 10Base-T implementation

McGraw-Hill
 Figure 13.13 10Base-F implementation

McGraw-Hill
 13.4 Fast Ethernet

 100 Mbps transmission rate

 same frame format, media access, and collision
detection rules as 10 Mbps Ethernet
 can combine 10 Mbps Ethernet and Fast Ethernet
on same network using a switch
 media: twisted pair (CAT 5) or fiber optic cable
(no coax)
 Star-wire topology
 Similar to 10BASE-T

CAT 3
CAT 5

McGraw-Hill
 Figure 13.19 Fast Ethernet topology

McGraw-Hill
 Figure 13.20 Fast Ethernet implementations

McGraw-Hill
 Fast Ethernet
and
Gigabit Ethernet

Networks: Fast Ethernet 29

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2000
 Fast Ethernet (100BASE-T)
How to achieve 100 Mbps capacity?

LLC
Data Link
Layer
MAC

Convergence Sublayer
Physical
MII Media Independent Interface Layer

Media Dependent Sublayer

Media Independent Interface provides three

choices.
Networks: Fast Ethernet 30
McGraw-Hill
 Fast Ethernet [IEEE 802.3u]
Three Choices

Figure 4-21.The original fast Ethernet

cabling.
* Concept facilitated by 10Mbps/100Mbps Adapter Cards

Networks: Fast Ethernet 31

McGraw-Hill
 100 BASE T

Networks: Fast Ethernet 32

McGraw-Hill
 Fast Ethernet Details
 UTP Cable has a 30 MHz limit
 Not feasible to use clock encoding (i.e., NO
Manchester encoding)
 Instead use bit encoding schemes with
sufficient transitions for receiver to
maintain clock synchronization.

Networks: Fast Ethernet 33

McGraw-Hill
 100 BASE T4

 Can use four separate twisted pairs of Cat 3

UTP
 Utilize three pair in both directions (at 33 1/3
Mbps) with other pair for carrier
sense/collision detection.
 Three-level ternary code is used 8B/6T.
Prior to transmission each set of 8 bits is
converted into 6 ternary symbols.
Networks: Fast Ethernet 34
McGraw-Hill
 100 BASE T4

 The signaling rate becomes

100 x 6/8
------------ = 25 MHz
3
 Three signal levels : +V, 0, -V
 Codewords are selected such that line is
d.c.balanced
 All codewords have a combined weight of 0
or 1.

Networks: Fast Ethernet 35

McGraw-Hill
 100 BASE T4
 36 = 729 possible codewords.
 Only 256 codewords are requires, hence they
are selected:
 To achieve d.c. balance
 Assuming all codewords have at least two signal
transitions within them (for receiver clock
synchronization).
 To solve d.c. wander, whenever a string of
codewords with +1 are sent, alternate
codewords (inverted before transmission) are
used.
 To reduce latency, ternary symbols are sent
staggered on the three lines.
Networks: Fast Ethernet 36
McGraw-Hill
 100 BASE T4

 Ethernet Interframe gap of 9.6

microseconds becomes 960 nanoseconds
in Fast Ethernet.
 100 m. max distance to hub; 200 meters
between stations.
 Maximum of two Class II repeaters.

Networks: Fast Ethernet 37

McGraw-Hill
 100 Base TX

 Uses two pair of twisted pair, one pair

for transmission and one pair for
reception.
 Uses either STP or Cat 5 UTP.
 Uses MTL-3 signaling scheme that
involves three voltages.
 Uses 4B/5B encoding.
 There is a guaranteed signal transition
at least every two bits.
Networks: Fast Ethernet 38
McGraw-Hill
 100 BASE FX

 Uses two optical fibers, one for

transmission and one for reception.
 Uses FDDI technology of converting 4B/5B
to NRZI code group streams into optical
signals.

Networks: Fast Ethernet 39

McGraw-Hill
 Fast Ethernet Repeaters and
Switches
 Class I Repeater – supports unlike physical
media segments (only one per collision
domain)
 Class II Repeater – limited to single physical
media type (there may be two repeaters per
collision domain)
 Switches – to improve performance can add
full-duplex and have autonegotiation for speed
mismatches.
Networks: Fast Ethernet 40
McGraw-Hill
 Collision Domains

Networks: Fast Ethernet 41

McGraw-Hill
 Networks: Fast Ethernet 42
McGraw-Hill
 Full Duplex Operation
 Traditional Ethernet is half duplex
 Either transmit or receive but not both simultaneously
 With full-duplex, station can transmit and receive data simultaneously
 With full duplex, Throughput (actual transmission rate) is doubled.
 10-Mbps Ethernet in full-duplex mode, theoretical transfer rate
becomes 20 Mbps
 100-Mbps Ethernet in full-duplex mode, theoretical transfer rate
becomes 200 Mbps
 Changes that should be made with any computer in order to operate
in Full-Duplex Mode
1) Attached stations must have full-duplex NIC cards
2) Must use two pairs of wire one pair for transmitting from host to
switch (inbound) and the other pair for transmitting from switch to
host (outbound)
3) Must use a switch as a central device not a hub
4) Devices must be connected point-to-point (dedicated) to the switch
 Each station constitutes separate collision domain
 CSMA/CD algorithm no longer needed (no collision)
 No limit on the segment length
McGraw-Hill
 Same 802.3 MAC frame format used
 Gigabit Ethernet History

 In February 1997 the Gigabit Ethernet Alliance

announced that IEEE802.3z Task Force met to
review the first draft of the Gigabit Ethernet
Standard
 According to IDC by the end of 1997 85% of all
network connections used Ethernet.
 Higher capacity Ethernet was appealing because
network managers can leverage their investment in
staff skills and training.
 1000 BASE X (IEEE802.3z) was ratified in June
1998.
Networks: Fast Ethernet 44
McGraw-Hill
 Figure 13.23 Gigabit Ethernet implementations

McGraw-Hill
 Gigabit Ethernet (1000 BASE

X) speeds of 1000 Mbps (i.e., one billion bits
Provides
per second capacity) for half-duplex and full-duplex
operation.
 Uses Ethernet frame format and MAC technology
 CSMA/CD access method with support for one repeater
per collision domain.
 Backward compatible with 10 BASE-T and 100 BASE-T.
 Uses 802.3 full-duplex Ethernet technology.
 Uses 802.3x flow control.
 All Gigabit Ethernet configurations are point-to-
point!
Networks: Fast Ethernet 46
McGraw-Hill
 Gigabit Ethernet Architecture Standard

Media Access Control (MAC)

full duplex and/or half duplex

Gigabit Media Independent Interface (GMII)

(optional)

1000 Base – X PHY

1000 Base T
8B/10B auto-negotiation PCS

1000 Base-LX 1000 Base-SX1000 Base-CX 1000 Base T

Fiber optic Fiber optic Copper PMA
transceiver transceiver transceiver transceiver
Single Mode or Multimode Shieled Unshielded twisted pair
Multimode Fiber Fiber Copper Cable IEEE 802.3ab

IEEE 802.3z

Source - IEEE
Networks: Fast Ethernet 47
McGraw-Hill
 Gigabit Ethernet Technology

Figure 4-23.Gigabit Ethernet cabling.

1000 BASE SX fiber - short wavelength
1000 BASE LX fiber - long wavelength
1000 BASE CX copper - shielded twisted pair
1000 BASE T copper - unshielded twisted pair
* Based on Fiber Channel physical signaling technology.

Networks: Fast Ethernet 48

McGraw-Hill
 Gigabit Ethernet (1000 BASE-
T)
LLC
Data Link
Layer
MAC

GMII Gigabit Media Independent Interface

Physical Layer

Media Dependent Interface

Medium

Networks: Fast Ethernet 49

McGraw-Hill
 Gigabit Media Independent
Interface
GMII
 Allows any physical layer to be used
with a given MAC.
 Namely, Fiber Channel physical layer
can be used with CSMA/CD.
 Permits both full-duplex and half-
duplex.

Networks: Fast Ethernet 50

McGraw-Hill
 1000 BASE SX
Short wavelength
 Supports duplex links up to 275 meters.
 770-860 nm range; 850 nm laser
wavelength
 (FC) Fiber Channel technology
 PCS (Physical Code Sublayer) includes
8B/10B encoding with 1.25 Gbps line.
 Only multimode fiber
 Cheaper than LX.
Networks: Fast Ethernet 51
McGraw-Hill
 8B/10B Encoder

Networks: Fast Ethernet 52

McGraw-Hill
 8B/10B Encoding Issues

• When the encoder has a choice for codewords, it always

chooses the codeword that moves in the direction of
balancing the number of 0s and 1s. This keeps the DC
component of the signal as low as possible.

Networks: Fast Ethernet 53

McGraw-Hill
 1000 BASE LX
Long wavelength

 Supports duplex links up to 550 meters.

 1270-1355 nm range; 1300 nm
wavelength using lasers.
 Fiber Channel technology
 PCS (Physical Code Sublayer) includes
8B/10B encoding with 1.25 Gbps line.
 Either single mode or multimode fiber.

Networks: Fast Ethernet 54

McGraw-Hill
 1000 BASE CX
‘Short haul’ copper jumpers
 Shielded twisted pair.

 25 meters or less typically within wiring

closet.
 PCS (Physical Code Sublayer) includes
8B/10B encoding with 1.25 Gbps line.
 Each link is composed of a separate
shielded twisted pair running in each
direction.

Networks: Fast Ethernet 55

McGraw-Hill
 1000 BASE T
Twisted Pair

 Four pairs of Category 5 UTP.

 IEEE 802.3ab ratified in June 1999.
 Category 5, 6 and 7 copper up to 100
meters.
 This requires extensive signal
processing.

Networks: Fast Ethernet 56

McGraw-Hill
 Gigabit Ethernet compared to
Fiber Channel
 Since Fiber Channel (FC) already existed,
the idea was to immediately leverage
physical layer of FC into Gigabit Ethernet.
 The difference is that fiber channel was
viewed as specialized for high-speed I/O
lines. Gigabit Ethernet is general purpose
and can be used as a high-capacity
switch.

Networks: Fast Ethernet 57

McGraw-Hill
 Gigabit Ethernet
 Viewed as LAN solution while ATM is WAN
solution.
 Gigabit Ethernet can be shared (hub) or
switched.
 Shared Hub
 Half duplex: CSMA/CD with MAC changes:
 Carrier Extension
 Frame Bursting
 Switch
 Full duplex: Buffered repeater called {Buffered
Distributor}
Networks: Fast Ethernet 58
McGraw-Hill
 Gigabit Ethernet

Figure 4-22. (a) A two-station Ethernet. (b) A multistation

Ethernet.

Networks: Fast Ethernet 59

McGraw-Hill
 Carrier Extension
Frame RRRRRRRRRRRRR

Carrier Extension
512 bytes

• For 10BaseT : 2.5 km max; slot time = 64 bytes

• For 1000BaseT: 200 m max; slot time = 512 bytes
• Carrier Extension :: continue transmitting control
characters [R] to fill collision interval.
• This permits minimum 64-byte frame to be handled.
• Control characters discarded at destination.
• For small frames net throughput is only slightly
better than Fast Ethernet.
Networks: Fast Ethernet Based on Raj Jain’s slide60
McGraw-Hill
 Frame Bursting
Frame Extension Frame Frame Frame

512 bytes

Frame burst

• Source sends out burst of frames without

relinquishing control of the network.
• Uses Ethernet Interframe gap filled with extension
bits (96 bits)
• Maximum frame burst is 8192 bytes
• Three times more throughput for small frames.

Networks: Fast Ethernet Based on Raj Jain’s slide 61

McGraw-Hill
 Buffered Distributor
Hub

 A buffered distributor is a new type of 802.3

hub where incoming frames are buffered in
FIFOs.
 CSMA/CD arbitration is inside the distributor to
transfer frames from an incoming FIFO to all
outgoing FIFOs.
 802.3x frame-based flow control is used to
handle congestion.
 All links are full-duplex. Based on Raj Jain slide
Networks: Fast Ethernet 62
McGraw-Hill
 Figure 13.18 Full-duplex switched Ethernet

McGraw-Hill
 Figure 13.17 Switched Ethernet

McGraw-Hill
 13.5 Gigabit Ethernet

 Speed 1Gpbs
 Minimum frame length is 512 bytes
 Operates in full/half duplex modes
mostly full duplex

McGraw-Hill
 Note

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.

McGraw-Hill
 Figure 13.22 Topologies of Gigabit Ethernet

McGraw-Hill
 10Gbps Ethernet
 Maximum link distances cover 300 m to 40 km
 Full-duplex mode only
 No CSMA/CD
 Uses optical fiber only

McGraw-Hill
 Overview

 What is 10 Gigabit Ethernet?

 Why 10 Gigabit Ethernet?
 Physical Layer Technologies
 10 Gigabit Ethernet in the LAN
 10 Gigabit Ethernet in the MAN
 10 Gigabit Ethernet in MAN over DWDM
 10 Gigabit Ethernet in the WAN

CS575 10 Gigabit Ethernet 69

McGraw-Hill
 What is 10 Gigabit Ethernet?
 An extension to 10, 100, and 1000 Mbps
IEEE 802.3 Ethernet
 Offering 10 Gigabits per second data rate

 An emerging IEEE 802.3ae Ethernet

standard
 The first draft of the standard is

expected to be completed by IEEE

802.3ae in 2001, with a full ratification of
this standard expected by the spring of
2002
CS575 10 Gigabit Ethernet 70
McGraw-Hill
10 Gigabit Ethernet will preserve many of
 What is 10 Gigabit Ethernet?
(continued)
 Initially, 10GbE will be used to provide
high-speed interconnection between large-
capacity switches in LAN environment
 As the need increases, 10GbE will be

deployed throughout the entire network

and will include server farm, backbone,
and campus-wide connectivity.
 In addition, the standard for 10 Gigabit

Ethernet is being developed with an

option
McGraw-Hill
CS575 for connection across
10 Gigabit Ethernet MAN and 71

WAN links at an expected data rate

 What is 10 Gigabit Ethernet?
(concluded)
 These connections will be over dark fiber,
dark wavelengths, or SONET/TDM
(Synchronous Optical Network) networks
 Such attachment to the optical WAN
backbone transport cloud, introduces the
concept of "Ethernet Everywhere."

CS575 10 Gigabit Ethernet 72

McGraw-Hill
 Why 10 Gigabit Ethernet?

 An explosive growth in both Internet and

intranet traffic due to:
 Increase in the number of network
connections
 Increase in the connection speed of each end-

station (e.g., 10Mbps users moving to

100Mbps, analog 56k users moving to DSL &
Cable modems)
 Increase in the deployment of bandwidth-

intensive applications such as high-quality

McGraw-Hill
video
CS575 10 Gigabit Ethernet 73

Increase in Web hosting and application

 Why 10 Gigabit Ethernet?
(concluded)
 Low cost of ownership – including both
acquisition and support costs
 Familiar management tools and common skills

base
 Ability to support new applications and data

types
 Flexibility in network design

 Leverages the installed base of more than 300

million Ethernet switch ports

CS575 10 Gigabit Ethernet 74
 Seamless integration of LAN, MAN, and
McGraw-Hill
 Physical Layer Technology

 Two families of physical interfaces are

defined by the IEEE 802.3ae Task Force:
 LAN PHY at 10,000 Mbps
 WAN PHY at a data rate compatible with OC-
192c/SDH VC-4-64c
 The IEEE 802.3ae Task Force has also
specified physical layer specifications that
support link distances of at least 65m over
MMF, 300m over installed MMF, as well as
over
CS575
SMF at 2km, 10km and 40km
10 Gigabit Ethernet 75
McGraw-Hill
Distance Objectives for IEEE 802.3ae
 Physical Layer Technology

Optical Transceiver for 10 Gigabit Ethernet

Source: 10 Gigabit Ethernet Alliance

CS575 10 Gigabit Ethernet 76

McGraw-Hill
 Physical Layer Technology
(concluded)
 The 10 Gigabit Ethernet Task Force has
identified an optional interface that matches
the data rate and protocol requirements of
SONET OC-192/SDH STM-64
 Therefore, 10 Gigabit Ethernet will be
compatible with SONET/SDH
 This enables direct attachment of packet-based

IP/Ethernet switches to the SONET/SDH and

time division multiplexed (TDM)
infrastructure
McGraw-Hill
CS575 10 Gigabit Ethernet 77

This feature is very important because it

 10 Gigabit Ethernet in the LAN

Source: 10 Gigabit Ethernet Alliance

CS575 10 Gigabit Ethernet 78

McGraw-Hill
 10 Gigabit Ethernet in the MAN

Source: 10 Gigabit Ethernet Alliance

CS575 10 Gigabit Ethernet 79

McGraw-Hill
 10 Gigabit Ethernet in MAN over
DWDM

Source: 10 Gigabit Ethernet Alliance

CS575 10 Gigabit Ethernet 80

McGraw-Hill
 10 Gigabit Ethernet in the WAN

Source: 10 Gigabit Ethernet Alliance

CS575 10 Gigabit Ethernet 81

McGraw-Hill
 How the WAN PHY Work

Source: 10 Gigabit Ethernet Alliance

CS575 10 Gigabit Ethernet 82

McGraw-Hill