Chapter 6:

The Link Layer (LAN)

 Layering in Internet protocol
stack
Applications
… built on ...
Application
Reliable (or unreliable) transport
Transport
… built on ...
Network
Best-effort global packet delivery
… built on ... Link

Best-effort local packet delivery Physical

… built on ...
Physical transfer of bits

Source: Scott Shenker (UC Berkeley): slide 7 at The Future of Networking, and the
Past of Protocols
https://www.youtube.com/watch?v=YHeyuD89n1Y&t=111s
Goals

 understand principles behind link

layer services:
• error detection, correction
• sharing a broadcast channel: multiple
access
• link layer addressing
• local area networks: Ethernet, VLANs
 instantiation, implementation of
various link layer technologies

Link Layer and LANs 6-3

Link layer, LANs: outline
6.1 introduction, 6.5 link virtualization:
services MPLS
6.2 error detection, 6.6 data center
correction networking
6.3 multiple access 6.7 a day in the life
protocols of a web request
6.4 LANs
• addressing, ARP
• Ethernet
• switches
• VLANS

Link Layer and LANs 6-4

 Link layer:
introduction
terminology:
 hosts and routers:
nodes
 communication
channels that connect
adjacent nodes along
communication path:
links
• wired links
• wireless links
• LANs
 layer-2 packet: frame,
data-link layer has responsibility of
encapsulates
transferring datagram datagram
from one node
to physically adjacent node over a link
Link Layer and LANs 6-5
Link layer: context
 datagram transferred transportation
by different link analogy:
protocols over different  trip from Princeton to
links: Lausanne
• e.g., Ethernet on • limo: Princeton to JFK
first link, frame relay • plane: JFK to Geneva
on intermediate • train: Geneva to
links, 802.11 on last Lausanne
link  tourist = datagram
 each link protocol  transport segment =
provides different communication link
services  transportation mode =
• e.g., may or may not link layer protocol
provide rdt over link  travel agent = routing
algorithm
Link Layer and LANs 6-6
Link layer services
 framing, link access:
• encapsulate datagram into frame, adding
header, trailer
• channel access if shared medium
• “MAC” addresses used in frame headers to
identify source, destination
• different from IP address!
 reliable delivery between adjacent
nodes
• we learned how to do this already (chapter
3)!
• seldom used on low bit-error link (fiber,
some twisted pair)
• wireless links: high error rates
• Q: why both link-level and end-end
reliability? Link Layer and LANs 6-7
Link layer services
(more)
 flow control:
• pacing between adjacent sending and receiving
nodes
 error detection:
• errors caused by signal attenuation, noise.
• receiver detects presence of errors:
• signals sender for retransmission or drops frame
 error correction:
• receiver identifies and corrects bit error(s) without
resorting to retransmission
 half-duplex and full-duplex
• with half duplex, nodes at both ends of link can
transmit, but not at same time

Link Layer and LANs 6-8

Where is the link layer
implemented?
Application

Transport

Network
Software
Link
Firmware Physical
/hardware
Where is the link layer
implemented?
 in each and every host
 link layer implemented
in “adaptor” (aka
network interface card
NIC) or on a chip application
• Ethernet card, transport
network cpu memory
802.11 card; link

Ethernet chipset host

• implements link, controller
bus
(e.g., PCI)
physical layer link
physical
 attaches into host’s physical
transmission
system buses
 combination of network adapter

hardware, software, card

firmware
Link Layer and LANs 6-10
Adaptors communicating

datagram datagram

controller controller

sending host receiving host

datagram

frame

 sending side:  receiving side

• encapsulates • looks for errors, rdt,
datagram in frame flow control, etc.
• adds error checking • extracts datagram,
bits, rdt, flow passes to upper
control, etc. layer at receiving
side
Link Layer and LANs 6-11
Link layer, LANs: outline
6.1 introduction, 6.5 link virtualization:
services MPLS
6.2 error detection, 6.6 data center
correction networking
6.3 multiple access 6.7 a day in the life
protocols of a web request
6.4 LANs
• addressing, ARP
• Ethernet
• switches
• VLANS

Link Layer and LANs 6-12

Error detection
EDC= Error Detection and Correction bits (redundancy)
D = Data protected by error checking, may include header fields

• Error detection never 100% accuracy!

• protocol may miss some errors, but rarely
• larger EDC field yields better detection and correction

otherwise

Link Layer and LANs 6-13

 Parity checking
single bit parity: two-dimensional bit parity:
 detect single bit  detect and correct single bit errors
errors

0 0

* Check out the online interactive exercises for more

examples: http://gaia.cs.umass.edu/kurose_ross/interactive/ Link Layer and LANs 6-14
Internet checksum (review)
goal: detect “errors” (e.g., flipped bits) in
transmitted packet (note: used at transport
layer only)
sender: receiver:
 treat segment  compute checksum of
contents as received segment
sequence of 16-bit  check if computed
integers checksum equals
 checksum: addition checksum field value:
(1’s complement • NO - error detected
sum) of segment • YES - no error
contents detected. But
 sender puts maybe errors
checksum value nonetheless?
into UDP checksum
field
Link Layer and LANs 6-15
Cyclic redundancy check (CRC)
 more powerful error-detection coding
 Input:

 data bits, D, as a binary number

 choose r+1 bit pattern (generator), G
 goal: choose r CRC bits, R, such
that <D,R> exactly divisible by G
(modulo 2)
 Check: receiver knows G, divides <D,R>
by G. If non-zero remainder: error
detected!
Modulo-2 Arithmetic
 Without carries in addition or borrows in
substraction
 Equivalent to bitwise XOR
• 1 XOR 1 = 0
• 0 XOR 1 = 1
• 1 XOR 0 = 1
• 0 XOR 0 = 0
 Subtraction equivalent to bitwise XOR
• 1011-0101 = ?111
• 1001-1101 = ?0100
0 Note: It is NOT the case 1001
1101
------------------------------------
borrows(1)
1100
CRC example
Note: 0 XOR R = R G D r=3
 Want: given D, G, r, to find R 101011
 <D,R> = D.2r XOR R is 1001 101110000
1001
exclusively divisible by G 101
 That is, 000
D.2r XOR R = nG 1010
 Equivalently 1001
110
D.2 = nG XOR R
r
000
 Equivalently 1100
1001
 if we divide D.2r by G, R
1010
want remainder R to 1001
satisfy: 011
R = remainder[ D 2
. r
]
G
CRC example: check
Note: 0 XOR R = R G D r=3
 Want: given D, G, r, to
101011
find R 1001 101110011
 <D,R> = D.2r XOR R is 1001
exclusively divisible by 101 R
G 000
 That is, 1010
1001
D.2r XOR R = nG 110
 Equivalently 000
D.2r = nG XOR R 1101
1001
 Equivalently 1001
 if we divide D.2r by G, 1001
want remainder D.2r R to 000
R = remainder[
satisfy: ]
G
FYI only: CRC Error Detection
Capability
A CRC of n-bit for divisor G(x) = (x+l)*p(x)
(i.e., admits factorization) can detect:
 All burst errors of length n.
 All burst errors affecting an odd number
of bits.
 All burst errors of length = n + 1 with
probability (2^(n-1) − l)/2^n − 1
 All burst errors of length > n + 1 with
probability (2^(n-1) − l)/2^n
• CRC-32 polynomial will detect all burst
errors of length greater than 33 with
probability (2^32 − l)/2^32 =
99.99999998% accuracy rate.
Data Link Layer 5-20
Link layer, LANs: outline
6.1 introduction, 6.5 link virtualization:
services MPLS
6.2 error detection, 6.6 data center
correction networking
6.3 multiple access 6.7 a day in the life
protocols of a web request
6.4 LANs
• addressing, ARP
• Ethernet
• switches
• VLANS

Link Layer and LANs 6-21

Multiple access links,
protocols
two types of “links”:
 point-to-point
• PPP for dial-up access
• point-to-point link between Ethernet switch, host
 broadcast (shared wire or medium)
• old-fashioned Ethernet
• upstream HFC
• 802.11 wireless LAN

shared wire (e.g., shared RF shared RF humans at a

cabled Ethernet) (e.g., 802.11 WiFi) (satellite) cocktail party
(shared air, acoustical)

Link Layer and LANs 6-22

Multiple access protocols
 single shared broadcast channel
 two or more simultaneous transmissions by nodes:
interference
• collision if node receives two or more signals at
the same time

multiple access protocol

 distributed algorithm that determines how nodes
share channel, i.e., determine when node can
transmit
 communication about channel sharing must use
channel itself!
• no out-of-band channel for coordination

Link Layer and LANs 6-23

An ideal multiple access
protocol
given: broadcast channel of rate R bps
desiderata:
1. when one node wants to transmit, it can send
at rate R.
2. when M nodes want to transmit, each can
send at average rate R/M
3. fully decentralized:
• no special node to coordinate transmissions
• no synchronization of clocks, slots
4. simple

Link Layer and LANs 6-24

MAC protocols: taxonomy
three broad classes:
 channel partitioning
• divide channel into smaller “pieces” (time slots,
frequency, code)
• allocate piece to node for exclusive use
 random access
• channel not divided, allow collisions
• “recover” from collisions
 “taking turns”
• nodes take turns, but nodes with more to send can
take longer turns

Link Layer and LANs 6-25

Channel partitioning MAC protocols:
TDMA
TDMA: time division multiple access
 access to channel in "rounds"
 each station gets fixed length slot
(length = packet transmission time) in
each round
 unused slots go idle
 example: 6-station LAN, 1,3,4 have
packets to send, slots 2,5,6 idle
6-slot 6-slot
frame frame
1 3 4 1 3 4

Link Layer and LANs 6-26

Channel partitioning MAC protocols:
FDMA
FDMA: frequency division multiple access
 channel spectrum divided into frequency bands
 each station assigned fixed frequency band
 unused transmission time in frequency bands go
idle
 example: 6-station LAN, 1,3,4 have packet to
send, frequency bands 2,5,6 idle
time
frequency bands

FDM cable

Link Layer and LANs 6-27

Random access protocols
 when node has packet to send
• transmit at full channel data rate R.
• no a priori coordination among nodes
 two or more transmitting nodes ➜
“collision”,
 random access MAC protocol specifies:
• how to detect collisions
• how to recover from collisions (e.g., via
delayed retransmissions)
 examples of random access MAC
protocols:
• slotted ALOHA
• ALOHA
• CSMA, CSMA/CD, CSMA/CA
Link Layer and LANs 6-28
Slotted ALOHA
assumptions: operation:
 all frames same size  when node obtains fresh
 time divided into frame, transmits in next
equal size slots (time slot
to transmit 1 frame) • if no collision: node
 nodes start to can send new frame
transmit only slot in next slot
beginning • if collision: node
 nodes are retransmits frame in
synchronized each subsequent slot
 if 2 or more nodes with prob. p until
transmit in slot, all success
nodes detect collision

Link Layer and LANs 6-29

Slotted ALOHA
node 1 1 1 1 1

node 2 2 2 2

node 3 3 3 3

C E C S E C E S S

Pros: Cons:
 single active node  collisions, wasting
can continuously slots
transmit at full rate  idle slots
of channel  nodes may be able
 highly decentralized:
to detect collision in
only slots in nodes
need to be in sync less than time to
 simple transmit packet
 clock
Link Layer and LANs 6-30
Slotted ALOHA: efficiency

efficiency: long-run  max efficiency: find

fraction of successful p* that maximizes
slots Np(1-p)N-1
(many nodes, all with  for many nodes, take
many frames to send) limit of Np*(1-p*)N-1
 suppose: N nodes as N goes to infinity,
with many frames to gives:
send, each transmits max efficiency = 1/e
in slot with = .37
at best:

!
probability p
 prob that given node channel
has success in a slot used for useful
= p(1-p)N-1 transmissions
 prob that any node 37%
has a success = of time!
Np(1-p)N-1 Link Layer and LANs 6-31
Pure (unslotted) ALOHA
 unslotted Aloha: simpler, no synchronization
 when frame first arrives
• transmit immediately
 collision probability increases:
• frame sent at t0 collides with other frames sent
in [t0-1,t0+1]

Link Layer and LANs 6-32

Pure ALOHA efficiency
P(success by given node) = P(node transmits) .
P(no other node transmits in [t0-
1,t0] .
P(no other node transmits in [t0-
1,t0]

= p . (1-p)N-1 . (1-p)N-1
= p . (1-p)2(N-1)

… choosing optimum p and then letting n

= 1/(2e) = .18
even worse than slotted Aloha!

Link Layer and LANs 6-33

CSMA (carrier sense multiple
access)
CSMA: listen before transmit:
if channel sensed idle: transmit
entire frame
 if channel sensed busy, defer
transmission

 human analogy: don’t interrupt

others!

Link Layer and LANs 6-34

CSMA collisions spatial layout of nodes

 collisions can still

occur: propagation
delay means two
nodes may not hear
each other’s
transmission
 collision: entire
packet transmission
time wasted
• distance &
propagation
delay play role in
in determining
collision
probability
Link Layer and LANs 6-35
CSMA/CD (collision
detection)
CSMA/CD: carrier sensing, deferral as in
CSMA
• collisions detected within short time
• colliding transmissions aborted, reducing
channel wastage
 collision detection:
• easy in wired LANs: measure signal strengths,
compare transmitted, received signals
• difficult in wireless LANs: received signal
strength overwhelmed by local transmission
strength
 human analogy: the polite
conversationalist
Link Layer and LANs 6-36
CSMA/CD (collision
detection)
spatial layout of nodes

Link Layer and LANs 6-37

Ethernet CSMA/CD
algorithm
1. NIC receives 4. If NIC detects
datagram from another
network layer, transmission while
creates frame transmitting, aborts
2. If NIC senses and sends jam
channel idle, starts signal
frame transmission. 5. After aborting, NIC
If NIC senses enters binary
channel busy, waits (exponential)
until channel idle, backoff:
then transmits. • after mth collision,
3. If NIC transmits NIC chooses K at
entire frame without random from
detecting another {0,1,2, …, 2m-1}.
transmission, NIC is NIC waits Link
K·512 bit
Layer and LANs 6-38
CSMA/CD efficiency
 Tprop = max prop delay between 2 nodes in LAN
 ttrans = time to transmit max-size frame

1
efficiency 
1  5t prop /ttrans
 efficiency goes to 1
• as tprop goes to 0
• as ttrans goes to infinity
 better performance than ALOHA: and simple,
cheap, decentralized!

Link Layer and LANs 6-39

“Taking turns” MAC
protocols
channel partitioning MAC protocols:
 share channel efficiently and fairly at high
load
 inefficient at low load: delay in channel
access, 1/N bandwidth allocated even if only
1 active node!
random access MAC protocols
 efficient at low load: single node can fully
utilize channel
 high load: collision overhead
“taking turns” protocols
look for best of both worlds!

Link Layer and LANs 6-40

“Taking turns” MAC
protocols
polling:
 master node
“invites” slave data
nodes to transmit poll
in turn
 typically used with master
“dumb” slave data
devices
 concerns:
• polling
overhead slaves
• latency
• single point of
failure (master)
Link Layer and LANs 6-41
“Taking turns” MAC
protocols
token passing:
T
 control token
passed from one
node to next
sequentially.
 token message (nothing
to send)
 concerns:
T
 token overhead
 latency
 single point of
failure (token)

data

Link Layer and LANs 6-42

Cable access
network
Internet frames, TV channels, control transmitted
downstream at different frequencies

cable headend

CMTS
…

splitter cable
cable modem … modem
termination system

ISP upstream Internet frames, TV control, transmitted

upstream at different frequencies in time slots

 multiple 40Mbps downstream (broadcast) channels

 single CMTS transmits into channels
 multiple 30 Mbps upstream channels
 multiple access: all users contend for certain
upstream channel time slots (others assigned)
Link Layer and LANs 6-43
Cable access
network
cable headend MAP frame for
Interval [t1, t2]

Downstream channel i
CMTS
Upstream channel j

t1 t2 Residences with cable modems

Minislots containing Assigned minislots containing cable modem

minislots request frames upstream data frames

DOCSIS: data over cable service interface

spec
 FDM over upstream, downstream frequency
channels
 TDM upstream: some slots assigned, some have
contention
• downstream MAP frame: assigns upstream
Link Layer and LANs 6-44
Summary of MAC protocols
 channel partitioning: TDMA, FDMA, CDMA
• share channel efficiently and fairly at high load
• inefficient at low load: delay in channel access, 1/N bandwidth
allocated even if only 1 active node!
 random access (dynamic), ALOHA, S-ALOHA, CSMA,
CSMA/CD (Ethernet), CSMA/CA (WiFi) (chapter 7)
• carrier sensing: easy in some technologies (wire), hard in
others (wireless)
• efficient at low load: single node can fully utilize channel
• high load: collision overhead
 taking turns, polling from central site, token passing
• Aim to take both merits,
• But, coordination (central or distributed), efficiency (overhead)
Link layer, LANs: outline
6.1 introduction, 6.5 link virtualization:
services MPLS
6.2 error detection, 6.6 data center
correction networking
6.3 multiple access 6.7 a day in the life
protocols of a web request
6.4 LANs
• addressing, ARP
• Ethernet
• switches
• VLANS

Link Layer and LANs 6-46

MAC addresses and ARP
 32-bit IP address:
• network-layer address for interface
• used for layer 3 (network layer) forwarding
 MAC (or LAN or physical or Ethernet)
address:
• function: used ‘locally” to get frame from one
interface to another physically-connected
interface (same network, in IP-addressing
sense)
• 48 bit MAC address (for most LANs) burned in
NIC ROM, also sometimes software settable
•hexadecimal
e.g.: 1A-2F-BB-76-09-AD
(base 16) notation
(each “numeral” represents 4 bits)

Link Layer and LANs 6-47

 LAN addresses and ARP
each adapter on LAN has unique LAN address

1A-2F-BB-76-09-AD

LAN
(wired or adapter
wireless)
71-65-F7-2B-08-53
58-23-D7-FA-20-B0

0C-C4-11-6F-E3-98

Link Layer and LANs 6-48

LAN addresses (more)
 MAC address allocation administered by
IEEE
 manufacturer buys portion of MAC
address space (to assure uniqueness)
 analogy:
• MAC address: like Social Security Number
• IP address: like postal address
 MAC flat address ➜ portability
• can move LAN card from one LAN to
another
 IP hierarchical address not portable
• address depends on IP subnet to which
node is attached
Link Layer and LANs 6-49
 ARP: address resolution
protocol
Question: how to determine
interface’s MAC address,
knowing its IP address? ARP table: each IP node
(host, router) on LAN
has table
137.196.7.78
• IP/MAC address
1A-2F-BB-76-09-AD
mappings for some
137.196.7.23
137.196.7.14 LAN nodes:
< IP address; MAC
address; TTL>
LAN
71-65-F7-2B-08-53
• TTL (Time To Live):
58-23-D7-FA-20-B0 time after which
address mapping
0C-C4-11-6F-E3-98 will be forgotten
137.196.7.88 (typically 20 min)
Link Layer and LANs 6-50
ARP protocol: same LAN
 A wants to send
datagram to B
• B’s MAC address not in  A caches (saves) IP-
A’s ARP table. to-MAC address pair
 A broadcasts ARP in its ARP table until
query packet, information becomes
containing B's IP old (times out)
address • soft state: information
• destination MAC that times out (goes
address = FF-FF-FF-FF- away) unless
FF-FF refreshed
• all nodes on LAN  ARP is “plug-and-
receive ARP query play”:
 B receives ARP • nodes create their ARP
packet, replies to A tables without
with its (B's) MAC intervention from net
administrator
address
• frame sent to A’s MAC Link Layer and LANs 6-51
 Addressing: routing to another
LAN
walkthrough: send datagram from A to B via R
 focus on addressing – at IP (datagram) and MAC
layer (frame)
 assume A knows B’s IP address
 assume A knows IP address of first hop router, R
(how?)
 assume A knows R’s MAC address (how?)

A B
R
111.111.111.111
222.222.222.222
74-29-9C-E8-FF-55
49-BD-D2-C7-56-2A
222.222.222.220
1A-23-F9-CD-06-9B

111.111.111.112 111.111.111.110 222.222.222.221

CC-49-DE-D0-AB-7D E6-E9-00-17-BB-4B 88-B2-2F-54-1A-0F

Link Layer and LANs 6-52

Addressing: routing to another
LAN
A creates IP datagram with IP source A, destination B
 A creates link-layer frame with R's MAC address as
destination address, frame contains A-to-B IP datagram
MAC src: 74-29-9C-E8-FF-55
MAC dest: E6-E9-00-17-BB-4B
IP src: 111.111.111.111
IP dest: 222.222.222.222

IP
Eth
Phy

A B
R
111.111.111.111
222.222.222.222
74-29-9C-E8-FF-55
49-BD-D2-C7-56-2A
222.222.222.220
1A-23-F9-CD-06-9B

111.111.111.112 111.111.111.110 222.222.222.221

CC-49-DE-D0-AB-7D E6-E9-00-17-BB-4B 88-B2-2F-54-1A-0F

Link Layer and LANs 6-53

Addressing: routing to another
LAN
frame sent from A to R
 frame received at R, datagram removed, passed up to IP

MAC src: 74-29-9C-E8-FF-55

MAC dest: E6-E9-00-17-BB-4B
IP src: 111.111.111.111
IP dest: 222.222.222.222
IP src: 111.111.111.111
IP dest: 222.222.222.222

IP IP
Eth Eth
Phy Phy

A B
R
111.111.111.111
222.222.222.222
74-29-9C-E8-FF-55
49-BD-D2-C7-56-2A
222.222.222.220
1A-23-F9-CD-06-9B

111.111.111.112 111.111.111.110 222.222.222.221

CC-49-DE-D0-AB-7D E6-E9-00-17-BB-4B 88-B2-2F-54-1A-0F

Link Layer and LANs 6-54

Addressing: routing to another
LAN
R forwards datagram with IP source A, destination B
 R creates link-layer frame with B's MAC address as
destination address, frame contains A-to-B IP datagram

MAC src: 1A-23-F9-CD-06-9B

MAC dest: 49-BD-D2-C7-56-2A
IP src: 111.111.111.111
IP dest: 222.222.222.222
IP
IP Eth
Eth Phy
Phy

A B
R
111.111.111.111
222.222.222.222
74-29-9C-E8-FF-55
49-BD-D2-C7-56-2A
222.222.222.220
1A-23-F9-CD-06-9B

111.111.111.112 111.111.111.110 222.222.222.221

CC-49-DE-D0-AB-7D E6-E9-00-17-BB-4B 88-B2-2F-54-1A-0F

Link Layer and LANs 6-55

Addressing: routing to another
LAN
R forwards datagram with IP source A, destination B
 R creates link-layer frame with B's MAC address as
destination address, frame contains A-to-B IP datagram

MAC src: 1A-23-F9-CD-06-9B

MAC dest: 49-BD-D2-C7-56-2A
IP src: 111.111.111.111
IP dest: 222.222.222.222
IP
IP Eth
Eth Phy
Phy

A B
R
111.111.111.111
222.222.222.222
74-29-9C-E8-FF-55
49-BD-D2-C7-56-2A
222.222.222.220
1A-23-F9-CD-06-9B

111.111.111.112 111.111.111.110 222.222.222.221

CC-49-DE-D0-AB-7D E6-E9-00-17-BB-4B 88-B2-2F-54-1A-0F

Link Layer and LANs 6-56

 Addressing: routing to another
LAN
 R forwards datagram with IP source A, destination B
 R creates link-layer frame with B's MAC address as dest,
frame contains A-to-B IP datagram
MAC src: 1A-23-F9-CD-06-9B
MAC dest: 49-BD-D2-C7-56-2A
IP src: 111.111.111.111
IP dest: 222.222.222.222

IP
Eth
Phy

A B
R
111.111.111.111
222.222.222.222
74-29-9C-E8-FF-55
49-BD-D2-C7-56-2A
222.222.222.220
1A-23-F9-CD-06-9B

111.111.111.112 111.111.111.110 222.222.222.221

CC-49-DE-D0-AB-7D E6-E9-00-17-BB-4B 88-B2-2F-54-1A-0F

* Check out the online interactive exercises for more

examples: http://gaia.cs.umass.edu/kurose_ross/interactive/
Link Layer and LANs 6-57
Link layer, LANs: outline
6.1 introduction, 6.5 link virtualization:
services MPLS
6.2 error detection, 6.6 data center
correction networking
6.3 multiple access 6.7 a day in the life
protocols of a web request
6.4 LANs
• addressing, ARP
• Ethernet
• switches
• VLANS

Link Layer and LANs 6-58

Ethernet
“dominant” wired LAN technology:
 single chip, multiple speeds (e.g., Broadcom
BCM5761)
 first widely used LAN technology
 simpler, cheap
 kept up with speed race: 10 Mbps – 10 Gbps

Metcalfe’s Ethernet sketch

Link Layer and LANs 6-59
Ethernet: physical topology
 bus: popular through mid 90s
• all nodes in same collision domain (can collide
with each other)
 star: prevails today
• active switch in center
• each “spoke” runs a (separate) Ethernet
protocol (nodes do not collide with each other)

switch
star
bus: coaxial cable
Link Layer and LANs 6-60
Ethernet frame structure
sending adapter encapsulates IP
datagram (or other network layer
protocol packet)type
in Ethernet frame
dest. source
preamble address address data CRC
(payload)

preamble:
 7 bytes with pattern 10101010 followed
by one byte with pattern 10101011
 used to synchronize receiver, sender
clock rates

Link Layer and LANs 6-61

Ethernet frame structure
(more)
 addresses: 6 byte source, destination MAC
addresses
• if adapter receives frame with matching
destination address, or with broadcast address
(e.g. ARP packet), it passes data in frame to
network layer protocol
• otherwise, adapter discards frame
 type: indicates higher layer protocol
(mostly IP but others possible, e.g., Novell
IPX, AppleTalk)
 CRC: cyclic redundancy
type check at receiver
• error detected:
preamble
dest. frame is
source dropped
data CRC
address address (payload)

Link Layer and LANs 6-62

Ethernet: unreliable,
connectionless
 connectionless: no handshaking between
sending and receiving NICs
 unreliable: receiving NIC doesn't send acks
or nacks to sending NIC
• data in dropped frames recovered only if
initial sender uses higher layer rdt (e.g.,
TCP), otherwise dropped data lost
 Ethernet’s MAC protocol: unslotted
CSMA/CD with binary backoff

Link Layer and LANs 6-63

802.3 Ethernet standards: link & physical
layers
 many different Ethernet standards
• common MAC protocol and frame format
• different speeds: 2 Mbps, 10 Mbps, 100
Mbps, 1Gbps, 10 Gbps, 40 Gbps
• different physical layer media: fiber, cable

MAC protocol
application and frame format
transport
network 100BASE-TX 100BASE-T2 100BASE-FX
link 100BASE-T4 100BASE-SX 100BASE-BX
physical

copper (twister fiber physical layer

pair) physical layer
Link Layer and LANs 6-64
Link layer, LANs: outline
6.1 introduction, 6.5 link virtualization:
services MPLS
6.2 error detection, 6.6 data center
correction networking
6.3 multiple access 6.7 a day in the life
protocols of a web request
6.4 LANs
• addressing, ARP
• Ethernet
• switches
• VLANS

Link Layer and LANs 6-65

Ethernet switch
 link-layer device: takes an active role
• store, forward Ethernet frames
• examine incoming frame’s MAC
address, selectively forward frame to
one-or-more outgoing links when frame
is to be forwarded on segment, uses
CSMA/CD to access segment
 transparent
• hosts are unaware of presence of
switches
 plug-and-play, self-learning
• switches do not need to be configured

Link Layer and LANs 6-66

Switch: multiple simultaneous
transmissions
 hosts have dedicated, A
direct connection to
switch C’ B
 switches buffer packets
6 1 2
 Ethernet protocol used on
each incoming link, but 5 4 3
no collisions; full duplex
• each link is its own B’ C
collision domain
 switching: A-to-A’ and B- A’
to-B’ can transmit switch with six interfaces
simultaneously, without (1,2,3,4,5,6)
collisions

Link Layer and LANs 6-67

Switch forwarding table
Q: how does switch know A
A’ reachable via interface B
C’
4, B’ reachable via
 A: each5?
interface switch has a 6 1 2
switch table, each
4
entry: 5 3

 (MAC address of host, B’ C

interface to reach host,
time stamp) A’
Q:how
looksare
likeentries
a routing table!
created, switch with six interfaces
maintained in switch (1,2,3,4,5,6)
table?
 something like a routing
protocol? Link Layer and LANs 6-68
Switch: self-learning Source: A
Dest: A’

 switch learns which A A A’

hosts can be reached B
C’
through which
interfaces 1
6 2
• when frame
received, switch 5 4 3
“learns” location of
sender: incoming B’ C
LAN segment
• records
A’
sender/location pair
in switch table MAC addr interface TTL
A 1 60 Switch table
(initially empty)

Link Layer and LANs 6-69

Switch: frame
filtering/forwarding
when frame received at switch:

1. record incoming link, MAC address of sending

host
2. index switch table using MAC destination
address
3. if entry found for destination
then {
if destination on segment from which frame
arrived
then drop frame
else forward frame on interface indicated
by entry
}
else flood /* forward on all interfaces except
Link Layer and LANs 6-70
Self-learning, forwarding: Source: A
example Dest: A’

A A A’
 frame destination,
B
A’, location flood C’

unknown:
 destination 1
A 6 2

locationselectively
known: A A’
5 4 3
send
B’ C
on just one link
A’ A

A’

MAC addr interface TTL

A 1 60 switch table
A’ 4 60 (initially empty)

Link Layer and LANs 6-71

Interconnecting switches
self-learning switches can be connected
together:
S4

S1
S3
A S2
F
D I
B C
G H
E

Q: sending from A to G - how does S1 know

to forward frame destined to G via S4 and
S 3?
 A: self learning! (works exactly the
same as in single-switch case!) Link Layer and LANs 6-72
Self-learning multi-switch
example
Suppose C sends frame to I, I responds to C

S4

S1
S3
A S2
F
D I
B C
G H
E

 Q: show switch tables and packet forwarding in

S1 , S 2 , S 3 , S 4

Link Layer and LANs 6-73

Institutional network
mail server
to external
network
router web server

IP subnet

Link Layer and LANs 6-74

Switches vs.
routers application
transport
both are store-and-
datagram network
forward: frame link
 routers: network-layer
physical link frame
devices (examine physical
network-layer
headers) switch
 switches: link-layer
devices (examine link- network datagram
layer headers) link frame
physical
both have forwarding application
tables: transport
 routers: compute network
tables using routing link
algorithms, IP physical
addresses
 switches: learn Link Layer and LANs 6-75
VLANs: motivation
consider:
 CS user moves office
to EE, but wants
connect to CS switch?
 single broadcast
domain:
• all layer-2
Computer
broadcast traffic
Science Electrical
Computer
Engineering
(ARP, DHCP,
Engineering unknown location
of destination MAC
address) must
cross entire LAN
• security/privacy,
efficiency issues
Link Layer and LANs 6-76
 port-based VLAN: switch ports
VLANs grouped (by switch
management software) so
that single physical switch
Virtual Local …… 1 7 9 15
Area Network 2 8 10 16

switch(es) supporting
VLAN capabilities can … …
be configured to Electrical Engineering Computer Science
define multiple virtual (VLAN ports 1-8) (VLAN ports 9-15)

LANS over single … operates as multiple virtual

physical LAN switches
infrastructure. 1 7 9 15
2 8 10 16

… …

Electrical Engineering Computer Science

(VLAN ports 1-8) (VLAN ports 9-16)

Link Layer and LANs 6-77

 Port-based VLAN
router
 traffic isolation: frames
to/from ports 1-8 can
only reach ports 1-8
• can also define VLAN
based on MAC addresses
of endpoints, rather than 1 7 9 15

2 8 10 16
switch port
 dynamic membership:
ports can be … …
dynamically assigned Electrical Engineering Computer Science
among VLANs (VLAN ports 1-8) (VLAN ports 9-15)

 forwarding between VLANS:

done via routing (just as with
separate switches)
• in practice vendors sell
combined switches plus routers

Link Layer and LANs 6-78

VLANS spanning multiple
switches
1 7 9 15 1 3 5 7

2 8 10 16 2 4 6 8

… …

Electrical Engineering Computer Science Ports 2,3,5 belong to EE VLAN

(VLAN ports 1-8) (VLAN ports 9-15) Ports 4,6,7,8 belong to CS VLAN

 trunk port: carries frames between VLANS defined

over multiple physical switches
• frames forwarded within VLAN between switches can’t be
vanilla 802.1 frames (must carry VLAN ID info)
• 802.1q protocol adds/removed additional header fields for
frames forwarded between trunk ports

Link Layer and LANs 6-79

802.1Q VLAN frame format
type

preamble dest. source data (payload) CRC

address address 802.1 frame

type
dest. source
preamble
address address
data (payload) CRC 802.1Q frame

2-byte Tag Protocol Identifier Recomputed

(value: 81-00) CRC

Tag Control Information (12 bit VLAN ID field,

3 bit priority field like IP TOS)

Link Layer and LANs 6-80

Link layer, LANs: outline
6.1 introduction, 6.5 link virtualization:
services MPLS
6.2 error detection, 6.6 data center
correction networking
6.3 multiple access 6.7 a day in the life
protocols of a web request
6.4 LANs
• addressing, ARP
• Ethernet
• switches
• VLANS

Link Layer and LANs 6-81

Data center networks
 10’s to 100’s of thousands of hosts, often
closely coupled, in close proximity:
• e-business (e.g. Amazon)
• content-servers (e.g., YouTube, Akamai, Apple,
Microsoft)
• search engines, data mining (e.g., Google)
 challenges:
 multiple applications,
each serving massive
numbers of clients
 managing/balancing
load, avoiding
processing, networking,
data bottlenecks Inside a 40-ft Microsoft container,
Chicago data center
Link Layer and LANs 6-82
Data center networks
load balancer: application-layer
routing
 receives external client requests
 directs workload within data center
Internet  returns results to external client
(hiding data center internals from
client)
Border router
Load Load
balancer Access router
balancer

Tier-1 switches
B

A C Tier-2 switches

TOR switches

Server racks

1 2 3 4 5 6 7 8
Link Layer and LANs 6-83
Data center networks
 rich interconnection among switches, racks:
• increased throughput between racks (multiple
routing paths possible)
• increased reliability via redundancy

Tier-1 switches

Tier-2 switches

TOR switches

Server racks

1 2 3 4 5 6 7 8
Link Layer and LANs 6-84
Link layer, LANs: outline
6.1 introduction, 6.5 link virtualization:
services MPLS
6.2 error detection, 6.6 data center
correction networking
6.3 multiple access 6.7 a day in the life
protocols of a web request
64 LANs
• addressing, ARP
• Ethernet
• switches
• VLANS

Link Layer and LANs 6-85

Synthesis: a day in the life of a web
request
 journey down protocol stack complete!
• application, transport, network, link
 putting-it-all-together: synthesis!
• goal: identify, review, understand protocols
(at all layers) involved in seemingly simple
scenario: requesting www page
• scenario: student attaches laptop to campus
network, requests/receives www.google.com

Link Layer and LANs 6-86

 A day in the life: scenario

browser DNS server

Comcast network
68.80.0.0/13

school network
68.80.2.0/24

web page

web server Google’s network

64.233.169.105 64.233.160.0/19

Link Layer and LANs 6-87

A day in the life… connecting to the
Internet
DHCP DHCP  connecting laptop
UDP
DHCP
DHCP IP
needs to get its own IP
DHCP Eth address, addr of first-
Phy hop router, addr of
DHCP DNS server: use DHCP
 DHCP request
encapsulated in UDP,
DHCP
DHCP
DHCP UDP
encapsulated in IP,
DHCP IP encapsulated in 802.3
DHCP Eth router Ethernet
Phy (runs DHCP)  Ethernet frame
broadcast (dest:
FFFFFFFFFFFF) on LAN,
received at router
running DHCP
 Ethernet server
demuxed to IP
demuxed, UDP
demuxed to DHCP
Link Layer and LANs 6-88
A day in the life… connecting to the
Internet
DHCP DHCP  DHCP server
DHCP UDP formulates DHCP ACK
DHCP IP containing client’s IP
DHCP Eth address, IP address of
Phy first-hop router for
client, name & IP
address of DNS
 encapsulation atserver
DHCP
DHCP DHCP server, frame
DHCP UDP forwarded (switch
DHCP IP learning) through LAN,
DHCP Eth router demultiplexing at
(runs DHCP)  DHCP
DHCP
Phy client client receives
DHCP ACK reply

Client now has IP address, knows name & addr of DNS

server, IP address of its first-hop router

Link Layer and LANs 6-89

A day in the life… ARP (before DNS,
before HTTP)
DNS DNS  before sending HTTP
DNS UDP request, need IP address of
DNS
ARP
IP www.google.com: DNS
ARP query Eth
Phy  DNS query created,
encapsulated in UDP,
encapsulated in IP,
ARP
encapsulated in Eth. To
ARP reply Eth
Phy send frame to router, need
 MAC address
ARP query of router
broadcast,
router
(runs DHCP)
interface:
received byARP
router, which
replies with ARP reply
giving MAC address of
router interface
 client now knows MAC
address of first hop router,
so can now send frame
containing DNS query
Link Layer and LANs 6-90
A day in the life… using DNSDNS
DNS UDP DNS server
DNS IP
DNS DNS DNS Eth
DNS UDP DNS Phy
DNS IP
DNS Eth
Phy
DNS
Comcast network
68.80.0.0/13

router
 IP datagram forwarded
(runs DHCP) from campus network into
 IP datagram containing Comcast network, routed
DNS query forwarded (tables created by RIP,
via LAN switch from OSPF, IS-IS and/or BGP
client to 1st hop router routing protocols)
 demuxed to DNS to DNS
server
server
 DNS server replies to
client with IP address
Link Layer and LANs 6-91
of www.google.com
A day in the life…TCP connection
carrying HTTP
HTTP
HTTP
SYNACK
SYN TCP
SYNACK
SYN IP
SYNACK
SYN Eth
Phy

 to send HTTP request,

client first opens TCP
router socket to web server
(runs DHCP)  TCP SYN segment (step 1
SYNACK
SYN TCP
SYNACK
SYN IP
in 3-way handshake)
SYNACK
SYN Eth inter-domain routed to
Phy  web server responds
with TCP SYNACK (step 2
web server in 3-way handshake)
64.233.169.105  TCP connection
established!
Link Layer and LANs 6-92
A day in the life… HTTP
request/reply
HTTP  web page finally (!!!)
HTTP HTTP
HTTP
HTTP TCP displayed
HTTP
HTTP IP
HTTP
HTTP Eth
Phy

 HTTP request sent into

TCP socket
router  IP datagram containing
HTTP (runs DHCP)
HTTP
HTTP TCP
HTTP request routed to
HTTP IP www.google.com
HTTP Eth  web server responds
Phy with HTTP reply
(containing web page)
web server
 IP datagram containing
64.233.169.105
HTTP reply routed back to
client
Link Layer and LANs 6-93
Chapter 6: Summary
 principles behind data link layer services:
• error detection, correction
• sharing a broadcast channel: multiple access
• link layer addressing
 instantiation and implementation of
various link layer technologies
• Ethernet
• switched LANS, VLANs
• virtualized networks as a link layer: MPLS
 synthesis: a day in the life of a web
request

Link Layer and LANs 6-94

Chapter 6: let’s take a
breath
 journey down protocol stack complete
(except PHY)
 solid understanding of networking
principles, practice
 ….. could stop here …. but lots of
interesting topics!
• wireless
• multimedia
• security

Link Layer and LANs 6-95