Machine Learning (ML) is a branch of Artificial Intelligence (AI)

Computer Networking: A
Top-Down Approach
8th
edition
Jim Kurose, Keith Ross
Pearson, 2020
Chapter 6
The Link Layer
and LANs
A note on the use of these PowerPoint slides:
We’re making these slides freely available to all (faculty, students,
readers). They’re in PowerPoint form so you see the animations; and
can add, modify, and delete slides (including this one) and slide content
to suit your needs. They obviously represent a lot of work on our part.
In return for use, we only ask the following:
 If you use these slides (e.g., in a class) that you mention their source
(after all, we’d like people to use our book!)
 If you post any slides on a www site, that you note that they are
adapted from (or perhaps identical to) our slides, and note our
copyright of this material.
For a revision history, see the slide note for this page.
Thanks and enjoy! JFK/KWR
All material copyright 1996-2023
J.F Kurose and K.W. Ross, All Rights Reserved

Link layer and LANs: our goals
understand principles
behind link layer services:
• error detection, correction
• sharing a broadcast channel:
multiple access
• link layer addressing
• local area networks:
Ethernet, VLANs
datacenter networks
 instantiation, implementation
of various link layer
technologies
Link Layer: 6-2

Link layer, LANs: roadmap
 a day in the life of a web
request
introduction
 error detection, correction
 multiple access protocols
 LANs
• addressing, ARP
• Ethernet
• switches
• VLANs
 link virtualization: MPLS
 data center networking
Link Layer: 6-3

Link layer: introduction
terminology:
 hosts, routers: nodes
 communication channels that
connect adjacent nodes along
communication path: links
• wired , wireless
• LANs
 layer-2 packet: frame,
encapsulates datagram
mobile network
enterprise
network
national or global ISP
datacenter
network
link layer has responsibility of
transferring datagram from one node
to physically adjacent node over a link
Link Layer 4

Link layer: context
 datagram transferred by
different link protocols over
different links:
• e.g., WiFi on first link,
Ethernet on next link
 each link protocol provides
different services
• e.g., may or may not provide
reliable data transfer over link
Link Layer 5

Transportation analogy
transportation analogy:
 trip from Princeton to Lausanne
• limo: Princeton to JFK
• plane: JFK to Geneva
• train: Geneva to Lausanne
 tourist = datagram
 transport segment =
communication link
 transportation mode = link-
layer protocol
 travel agent = routing algorithm
Princeton
JFK
Geneva Lausanne
Link Layer 6

Link layer: services
 framing, link access:
• encapsulate datagram into frame, adding
header, trailer
• channel access if shared medium
• “MAC” addresses in frame headers identify
source, destination (different from IP
address!)
 reliable delivery between adjacent nodes
• we already know how to do this!
• seldom used on low bit-error links
• wireless links: high error rates
• Q: why both link-level and end-end
reliability?
…
…
Cable access
cellular
WiFi
Ethernet LANs
Link Layer 7

Link layer: services (more)
 flow control:
• pacing between adjacent sending and receiving
nodes
 error detection:
• errors caused by signal attenuation, noise.
• receiver detects errors, signals retransmission, or
drops frame
 error correction:
• receiver identifies and corrects bit error(s)
without retransmission
 half-duplex and full-duplex:
• with half duplex, nodes at both ends of link can
transmit, but not at same time
…
…
Cable access
cellular
WiFi
Ethernet LANs
Link Layer 8

Host link-layer implementation
 in each-and-every host
 link layer implemented on-chip or
in network interface card (NIC)
• implements link, physical layer
 attaches into host’s system buses
 combination of hardware,
software, firmware
controller
physical
cpu memory
host bus
(e.g., PCI)
network interface
application
transport
network
link
link
physical
Link Layer 9

controller
physical
memory CPU
Interfaces communicating
controller
physical
cpu memory
application
transport
network
link
link
physical
application
transport
network
link
link
physical
sending side:
 encapsulates datagram in frame
 adds error checking bits, reliable data
transfer, flow control, etc.
receiving side:
 looks for errors, reliable data
transfer, flow control, etc.
 extracts datagram, passes to
upper layer at receiving side
linkh
linkh datagram
datagram
datagram
Link Layer 10

request
 introduction
error detection, correction
 LANs
• addressing, ARP
• Ethernet
• switches
• VLANs
Link Layer: 6-11

Error detection
EDC: error detection and correction bits (e.g., redundancy)
D: data protected by error checking, may include header fields
Error detection not 100%
reliable!
 protocol may miss
some errors, but rarely
 larger EDC field yields
better detection and
correction
datagram
D EDC
d data bits
bit-error prone link
D’ EDC’
all
bits in D’
OK
?
N
detected
error
otherwise
datagram
Link Layer 12

Parity checking
single bit parity:
 detect single bit errors
0111000110101011 1
parity bit
d data bits
Can detect and correct errors
(without retransmission!)
 two-dimensional parity: detect
and correct single bit errors
d1,1
d2,1
di,1
. . .
d1,j+1
d2,j+1
di,j+1
. . .
. . .
d1,j
d2,j
di,j
. . .
di+1,1 di+1,j+1
di+1,j
. . .
. . .
. . .
. . .
row parity
column
parity
1 0 1 0 1
0 1 1 1 0
1 0 1 0 1
1 1 1 1 0
1
0
1
0
no errors:
parity
error
parity
error
0 1 1 1 0 1
1 0 1 0 1 1
1 0 1 1 0 0
1 0 1 0 1 0
detected
and
correctable
single-bit
error:
Even/odd parity: set parity bit so
there is an even/odd number of 1’s
* Check out the online interactive exercises for more examples: http://gaia.cs.umass.edu/kurose_ross/interactive/
At receiver:
 compute parity of d received
bits
 compare with received parity bit
– if different than error detected

Internet checksum (review, see section 3.3)
sender:
 treat contents of UDP
segment (including UDP header
fields and IP addresses) as
sequence of 16-bit integers
 checksum: addition (one’s
complement sum) of segment
content
 checksum value put into
UDP checksum field
receiver:
 compute checksum of received
segment
 check if computed checksum equals
checksum field value:
• not equal - error detected
• equal - no error detected. But maybe
errors nonetheless? More later ….
Goal: detect errors (i.e., flipped bits) in transmitted segment
Link Layer 14

Cyclic Redundancy Check (CRC)
 more powerful error-detection coding
 D: data bits (given, think of these as a binary number)
 G: bit pattern (generator), of r+1 bits (given, specified in CRC standard)
sender: compute r CRC bits, R, such that <D,R> exactly divisible by G (mod 2)
• receiver knows G, divides <D,R> by G. If non-zero remainder: error detected!
• can detect all burst errors less than r+1 bits
• widely used in practice (Ethernet, 802.11 WiFi)
<D,R> = D 2r
XOR R
* formula for these bits
r CRC bits
d data bits
D R bits to send
Link Layer 15

Cyclic Redundancy Check (CRC): example
Sender wants to compute R
such that:
D . 2r
XOR R = nG
* Check out the online interactive exercises for more examples: http://gaia.cs.umass.edu/kurose_ross/interactive/
D.2r
G
R = remainder [ ]
... or equivalently (XOR R both sides):
D . 2r
= nG XOR R
... which says:
if we divide D . 2r
by G, we
want remainder R to satisfy:
1 0 0 1
1 0 1 0
1 0 1
0 0 0
1 0 0 1
1 0 0 1
1 0 0 1
0 0 0
1 1 0
1 1 0 0
1 0 1 0
0 1 1
0 1 1
D
R
1 0 0 1
G
0 0 0
1 0 1 1 1 0
2r
(here, r=3)
*
1 0 1
algorithm for
computing R
Link Layer 16

request
 introduction
multiple access protocols
 LANs
• addressing, ARP
• Ethernet
• switches
• VLANs
Link Layer: 6-17

Multiple access links, protocols
two types of “links”:
 point-to-point
• point-to-point link between Ethernet switch, host
• PPP for dial-up access
 broadcast (shared wire or medium)
• old-school Ethernet
• upstream HFC in cable-based access network
• 802.11 wireless LAN, 4G/4G. satellite
shared radio: satellite humans at a cocktail party
(shared air, acoustical)
shared radio: WiFi
shared wire (e.g.,
cabled Ethernet)
shared radio: 4G/5G
Link Layer 18

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
 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
multiple access protocol
Link Layer 19

An ideal multiple access protocol
given: multiple access channel (MAC) 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 20

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 21

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
1 3 4 1 3 4
6-slot
frame
6-slot
frame
Link Layer 22

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
frequency
bands
time
FDM cable
Link Layer 23

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 protocol specifies:
• how to detect collisions
• how to recover from collisions (e.g., via delayed retransmissions)
 examples of random access MAC protocols:
• ALOHA, slotted ALOHA
• CSMA, CSMA/CD, CSMA/CA
Link Layer 24

Slotted ALOHA
assumptions:
 all frames same size
 time divided into equal size
slots (time to transmit 1 frame)
 nodes start to transmit only
slot beginning
 nodes are synchronized
 if 2 or more nodes transmit in
slot, all nodes detect collision
operation:
 when node obtains fresh
frame, transmits in next slot
• if no collision: node can send
new frame in next slot
• if collision: node retransmits
frame in each subsequent
slot with probability p until
success
randomization – why?
t0 t0+1
Link Layer 25

Slotted ALOHA
Pros:
 single active node can
continuously transmit at full rate
of channel
 highly decentralized: only slots in
nodes need to be in sync
 simple
Cons:
 collisions, wasting slots
 idle slots
 nodes may be able to detect collision in
less than time to transmit packet
 clock synchronization
1 1 1 1
2
3
2 2
3 3
node 1
node 2
node 3
C C C
S S S
E E E
C: collision
S: success
E: empty
Link Layer 26

efficiency: long-run fraction of successful slots (many nodes, all with
many frames to send)
 suppose: N nodes with many frames to send, each transmits in slot with
probability p
• prob that given node has success in a slot = p(1-p)N-1
• prob that any node has a success = Np(1-p)N-1
• max efficiency: find p* that maximizes Np(1-p)N-1
• for many nodes, take limit of Np*(1-p*)N-1
as N goes to infinity, gives:
max efficiency = 1/e = .37
at best: channel used for useful transmissions 37% of time!
Slotted ALOHA: efficiency
Link Layer 27

CSMA (carrier sense multiple access)
simple CSMA: listen before transmit:
• if channel sensed idle: transmit entire frame
• if channel sensed busy: defer transmission
 human analogy: don’t interrupt others!
CSMA/CD: CSMA with collision detection
• collisions detected within short time
• colliding transmissions aborted, reducing channel wastage
• collision detection easy in wired, difficult with wireless
 human analogy: the polite conversationalist
Link Layer 29

CSMA: collisions
 collisions can still occur with
carrier sensing:
• propagation delay means two nodes
may not hear each other’s just-
started transmission
 collision: entire packet
transmission time wasted
• distance & propagation delay play
role in in determining collision
probability
spatial layout of nodes
Link Layer 30

CSMA/CD:
 CSMA/CD reduces the amount of
time wasted in collisions
• transmission aborted on collision
detection
spatial layout of nodes
Link Layer 31

Ethernet CSMA/CD algorithm
1. Ethernet receives datagram from network layer, creates frame
2. If Ethernet senses channel:
if idle: start frame transmission.
if busy: wait until channel idle, then transmit
3. If entire frame transmitted without collision - done!
4. If another transmission detected while sending: abort, send jam signal
5. After aborting, enter binary (exponential) backoff:
• after mth collision, chooses K at random from {0,1,2, …, 2m
-1}.
Ethernet waits K·512 bit times, returns to Step 2
• more collisions: longer backoff interval
Link Layer 32

“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 34

polling:
 centralized controller “invites”
other nodes to transmit in turn
 typically used with “dumb”
devices
 concerns:
• polling overhead
• latency
• single point of failure (master)
• Bluetooth uses polling
centralized
controller
client devices
poll
data
data
Link Layer 35

token passing:
 control token message
explicitly passed from one node
to next, sequentially
 transmit while holding token
 concerns:
• token overhead
• latency
• single point of failure
(token)
T
data
(nothing
to send)
T
Link Layer 36

Cable access network: FDM, TDM and random access!
Link Layer: 6-37
cable headend
CMTS
ISP cable modem
termination system
cable
modem
splitter
…
…
Internet frames, TV channels, control transmitted
downstream at different frequencies
 multiple downstream (broadcast) FDM channels: up to 1.6 Gbps/channel
 single CMTS transmits into channels
 multiple upstream channels (up to 1 Gbps/channel)
 multiple access: all users contend (random access) for certain upstream
channel time slots; others assigned TDM

Cable access network:
Link Layer: 6-38
DOCSIS: data over cable service interface specification
 FDM over upstream, downstream frequency channels
 TDM upstream: some slots assigned, some have contention
• downstream MAP frame: assigns upstream slots
• request for upstream slots (and data) transmitted random access (binary
backoff) in selected slots
Residences with cable modems
Downstream channel i
Upstream channel j
MAP frame for
Interval [t1, t2]
t1 t2
Assigned minislots containing cable modem
upstream data frames
Minislots containing
minislots request frames
cable headend
CMTS

Summary of MAC protocols
Link Layer: 6-39
 channel partitioning, by time, frequency or code
• Time Division, Frequency Division
 random access (dynamic),
• ALOHA, S-ALOHA, CSMA, CSMA/CD
• carrier sensing: easy in some technologies (wire), hard in others
(wireless)
• CSMA/CD used in Ethernet
• CSMA/CA used in 802.11
 taking turns
• polling from central site, token passing
• Bluetooth, FDDI, token ring

request
 introduction
 LANs
• addressing, ARP
• Ethernet
• switches
• VLANs
Link Layer: 6-40

MAC addresses
Link Layer: 6-41
 32-bit IP address:
• network-layer address for interface
• used for layer 3 (network layer) forwarding
• e.g.: 128.119.40.136
 MAC (or LAN or physical or Ethernet) address:
• function: used “locally” to get frame from one interface to another
physically-connected interface (same subnet, in IP-addressing sense)
• 48-bit MAC address (for most LANs) burned in NIC ROM, also
sometimes software settable
hexadecimal (base 16) notation
(each “numeral” represents 4 bits)
• e.g.: 1A-2F-BB-76-09-AD

MAC addresses
Link Layer: 6-42
each interface on LAN
 has unique 48-bit MAC address
 has a locally unique 32-bit IP address (as we’ve seen)
1A-2F-BB-76-09-AD
58-23-D7-FA-20-B0
0C-C4-11-6F-E3-98
71-65-F7-2B-08-53
LAN
(wired or wireless)
137.196.7/24
137.196.7.78
137.196.7.14
137.196.7.88
137.196.7.23

MAC addresses
Link Layer: 6-43
 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 interface from one LAN to another
• recall IP address not portable: depends on IP subnet to which
node is attached

ARP: address resolution protocol
Link Layer: 6-44
ARP table: each IP node (host,
router) on LAN has table
Question: how to determine interface’s MAC address, knowing its IP
address?
1A-2F-BB-76-09-AD
58-23-D7-FA-20-B0
0C-C4-11-6F-E3-98
71-65-F7-2B-08-53
LAN
137.196.7.78
137.196.7.14
137.196.7.88
137.196.7.23
ARP
ARP
ARP
ARP
• IP/MAC address mappings for
some LAN nodes:
< IP address; MAC address; TTL>
• TTL (Time To Live): time after
which address mapping will be
forgotten (typically 20 min)

ARP protocol in action
Link Layer: 6-45
58-23-D7-FA-20-B0
137.196.7.14
B
C
D
TTL
71-65-F7-2B-08-53
137.196.7.23
A
ARP table in A
IP addr MAC addr TTL
example: A wants to send datagram to B
• B’s MAC address not in A’s ARP table, so A uses ARP to find B’s MAC address
A broadcasts ARP query, containing B's IP addr
• destination MAC address = FF-FF-FF-FF-FF-FF
• all nodes on LAN receive ARP query
1 Source MAC: 71-65-F7-2B-08-53
Source IP: 137.196.7.23
Target IP address: 137.196.7.14
…
1
Ethernet frame (sent to FF-FF-FF-FF-FF-FF)

Link Layer: 6-46
58-23-D7-FA-20-B0
137.196.7.14
B
C
D
TTL
71-65-F7-2B-08-53
137.196.7.23
A
ARP table in A
B replies to A with ARP response,
giving its MAC address
2
Target IP address: 137.196.7.14
Target MAC address:
58-23-D7-FA-20-B0
…
2
ARP message into Ethernet frame
(sent to 71-65-F7-2B-08-53)

Link Layer: 6-47
58-23-D7-FA-20-B0
137.196.7.14
B
C
D
TTL
71-65-F7-2B-08-53
137.196.7.23
A
ARP table in A
A receives B’s reply, adds B entry
into its local ARP table
3
137.196.
7.14
58-23-D7-FA-20-B0 500

Routing to another subnet: addressing
Link Layer: 6-48
walkthrough: sending a datagram from A to B via R
 focus on addressing – at IP (datagram) and MAC layer (frame) levels
R
A B
1A-23-F9-CD-06-9B
222.222.222.220
111.111.111.110
E6-E9-00-17-BB-4B
CC-49-DE-D0-AB-7D
111.111.111.112
111.111.111.111
74-29-9C-E8-FF-55 222.222.222.222
49-BD-D2-C7-56-2A
222.222.222.221
88-B2-2F-54-1A-0F
 assume that:
• A knows B’s IP address
• A knows IP address of first hop router, R (how?)
• A knows R’s MAC address (how?)

Link Layer: 6-49
R
1A-23-F9-CD-06-9B
222.222.222.220
111.111.111.110
E6-E9-00-17-BB-4B
CC-49-DE-D0-AB-7D
111.111.111.112
111.111.111.111
74-29-9C-E8-FF-55
A
222.222.222.222
49-BD-D2-C7-56-2A
222.222.222.221
88-B2-2F-54-1A-0F
B
IP
Eth
Phy
IP src: 111.111.111.111
IP dest: 222.222.222.222
 A creates IP datagram with IP source A, destination B
 A creates link-layer frame containing A-to-B IP datagram
• R's MAC address is frame’s destination
MAC src: 74-29-9C-E8-FF-55
MAC dest: E6-E9-00-17-BB-4B

Link Layer: 6-50
R
1A-23-F9-CD-06-9B
222.222.222.220
111.111.111.110
E6-E9-00-17-BB-4B
CC-49-DE-D0-AB-7D
111.111.111.112
111.111.111.111
74-29-9C-E8-FF-55
A
222.222.222.222
49-BD-D2-C7-56-2A
222.222.222.221
88-B2-2F-54-1A-0F
B
IP
Eth
Phy
 frame sent from A to R
IP
Eth
Phy
 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

Link Layer: 6-51
R
1A-23-F9-CD-06-9B
222.222.222.220
111.111.111.110
E6-E9-00-17-BB-4B
CC-49-DE-D0-AB-7D
111.111.111.112
111.111.111.111
74-29-9C-E8-FF-55
A
222.222.222.222
49-BD-D2-C7-56-2A
222.222.222.221
88-B2-2F-54-1A-0F
B
IP src: 111.111.111.111
IP dest: 222.222.222.222
MAC src: 1A-23-F9-CD-06-9B
MAC dest: 49-BD-D2-C7-56-2A
 R determines outgoing interface, passes datagram with IP source A, destination B
to link layer
 R creates link-layer frame containing A-to-B IP datagram. Frame destination address:
B's MAC address
IP
Eth
Phy

Link Layer: 6-52
R
1A-23-F9-CD-06-9B
222.222.222.220
111.111.111.110
E6-E9-00-17-BB-4B
CC-49-DE-D0-AB-7D
111.111.111.112
111.111.111.111
74-29-9C-E8-FF-55
A
222.222.222.222
49-BD-D2-C7-56-2A
222.222.222.221
88-B2-2F-54-1A-0F
B
IP
Eth
Phy
IP
Eth
Phy
IP src: 111.111.111.111
IP dest: 222.222.222.222
MAC src: 1A-23-F9-CD-06-9B
MAC dest: 49-BD-D2-C7-56-2A
 transmits link-layer frame
 R determines outgoing interface, passes datagram with IP source A, destination B
to link layer
 R creates link-layer frame containing A-to-B IP datagram. Frame destination address:
B's MAC address

Link Layer: 6-53
R
1A-23-F9-CD-06-9B
222.222.222.220
111.111.111.110
E6-E9-00-17-BB-4B
CC-49-DE-D0-AB-7D
111.111.111.112
111.111.111.111
74-29-9C-E8-FF-55
A
222.222.222.222
49-BD-D2-C7-56-2A
222.222.222.221
88-B2-2F-54-1A-0F
B
IP
Eth
Phy
IP
Eth
Phy
 B receives frame, extracts IP datagram destination B
 B passes datagram up protocol stack to IP
IP src: 111.111.111.111
IP dest: 222.222.222.222

request
 introduction
 LANs
• addressing, ARP
• Ethernet
• switches
• VLANs
Link Layer: 6-54

Ethernet
Link Layer: 6-55
“dominant” wired LAN technology:
 first widely used LAN technology
 simpler, cheap
 kept up with speed race: 10 Mbps – 400 Gbps
 single chip, multiple speeds (e.g., Broadcom BCM5761)
Metcalfe’s Ethernet
sketch
https://www.uspto.gov/learning-and-resources/journeys-innovation/audio-stories/defying-doubters
Bob Metcalfe: Ethernet co-inventor,
2022 ACM Turing Award recipient

Ethernet: physical topology
Link Layer: 6-56
 bus: popular through mid 90s
• all nodes in same collision domain (can collide with each other)
bus: coaxial cable switched
 switched: prevails today
• active link-layer 2 switch in center
• each “spoke” runs a (separate) Ethernet protocol (nodes do not collide with
each other)

Ethernet frame structure
Link Layer: 6-57
sending interface encapsulates IP datagram (or other network layer
protocol packet) in Ethernet frame
dest.
address
source
address data (payload) CRC
preamble
type
preamble:
 used to synchronize receiver, sender clock rates
 7 bytes of 10101010 followed by one byte of 10101011

Ethernet frame structure (more)
Link Layer: 6-58
dest.
address
source
preamble
type
 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
• used to demultiplex up at receiver
 CRC: cyclic redundancy check at receiver
• error detected: frame is dropped

Ethernet: unreliable, connectionless
Link Layer: 6-59
connectionless: no handshaking between sending and
receiving NICs
unreliable: receiving NIC doesn’t send ACKs or NAKs 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

802.3 Ethernet standards: link & physical layers
Link Layer: 6-60
• different physical layer media: fiber, cable
application
transport
network
link
physical
MAC protocol
and frame format
100BASE-TX
100BASE-T4
100BASE-FX
100BASE-T2
100BASE-SX 100BASE-BX
fiber physical layer
copper (twister pair) physical layer
 many different Ethernet standards
• common MAC protocol and frame format
• different speeds: 2 Mbps, ... 100 Mbps, 1Gbps, 10 Gbps, 40 Gbps, 80 Gbps

request
 introduction
 LANs
• addressing, ARP
• Ethernet
• switches
• VLANs
Link Layer: 6-61

Ethernet switch
Link Layer: 6-62
 Switch is a link-layer device: takes an active role
• store, forward Ethernet (or other type of) 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 unaware of presence of switches
 plug-and-play, self-learning
• switches do not need to be configured

Switch: multiple simultaneous transmissions
Link Layer: 6-63
switch with six
interfaces (1,2,3,4,5,6)
A
A’
B
B’ C
C’
1 2
3
4
5
6
 hosts have dedicated, direct
connection to switch
 switches buffer packets
 Ethernet protocol used on each
incoming link, so:
• no collisions; full duplex
• each link is its own collision domain
 switching: A-to-A’ and B-to-B’ can transmit
simultaneously, without collisions

Switch: multiple simultaneous transmissions
Link Layer: 6-64
switch with six
interfaces (1,2,3,4,5,6)
A
A’
B
B’ C
C’
1 2
3
4
5
6
 hosts have dedicated, direct
connection to switch
 switches buffer packets
 Ethernet protocol used on each
incoming link, so:
• no collisions; full duplex
• each link is its own collision domain
 switching: A-to-A’ and B-to-B’ can transmit
simultaneously, without collisions
• but A-to-A’ and C to A’ can not happen
simultaneously

Switch forwarding table
Link Layer: 6-65
A
A’
B
B’ C
C’
1 2
3
4
5
6
Q: how does switch know A’ reachable via
interface 4, B’ reachable via interface 5?
A: each switch has a switch table, each
entry:
 (MAC address of host, interface to reach
host, time stamp)
 looks like a routing table!
Q: how are entries created, maintained
in switch table?
 something like a routing protocol?

Switch: self-learning
Link Layer: 6-66
A
A’
B
B’ C
C’
1 2
3
4
5
6
switch learns which hosts
can be reached through
which interfaces
A A’
Source: A
Dest: A’
MAC addr interface TTL
Switch table
(initially empty)
A 1 60
• when frame received, switch
“learns” location of sender:
incoming LAN segment
• records sender/location pair
in switch table

Switch: frame filtering/forwarding
Link Layer: 6-67
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 arriving interface */

A
A’
B
B’ C
C’
1 2
3
4
5
6
Self-learning, forwarding: example
Link Layer: 6-68
A A’
Source: A
Dest: A’
MAC addr interface TTL
switch table
(initially empty)
A 1 60
A A’
A A’
A A’
A A’
A A’
A’ A
A’ 4 60
 frame destination, A’,
location unknown: flood
 destination A location
known: selectively send
on just one link

Interconnecting switches
Link Layer: 6-69
self-learning switches can be connected together:
Q: sending from A to G - how does S1 know to forward frame destined to
G via S4 and S3?
 A: self learning! (works exactly the same as in single-switch case!)
A
B
S1
C D
E
F
S2
S4
S3
H
I
G

Self-learning multi-switch example
Link Layer: 6-70
Suppose C sends frame to I, I responds to C
Q: show switch tables and packet forwarding in S1, S2, S3, S4
A
B
S1
C D
E
F
S2
S4
S3
H
I
G

...
Agg2
...
Agg1
building
closets
UMass Campus Network- Detail
to off campus
Wireless
Controller
...
Wireless
Controller
border border
core
Core
...
Agg3
...
Agg4 firewall data center
WiFi
UMass network:
 4 firewalls
 10 routers
 2000+ network switches
 6000 wireless access
points
 30000 active wired
network jacks
 55000 active end-user
wireless devices
… all built,
operated,
maintained by
~15 people

...
Agg2
...
Agg1
building
closets
UMass Campus Network- Detail
to off campus
Wireless
Controller
...
Wireless
Controller
border border
core
Core
...
Agg3
...
Agg4 firewall data center
WiFi
eBGP
iBGP
IS-IS
IS-IS
Ethernet
layer-2
switching
intra-domain
routing
inter-domain
routing
Protocols Link Speeds
10G;
100G pending
40G & 100G
40G
10G & 1G

Switches vs. routers
Link Layer: 6-73
application
transport
network
link
physical
network
link
physical
link
physical
switch
datagram
application
transport
network
link
physical
frame
frame
frame
datagram
6-73
both are store-and-forward:
 routers: network-layer devices (examine
network-layer headers)
 switches: link-layer devices (examine
link-layer headers)
both have forwarding tables:
 routers: compute tables using routing
algorithms, IP addresses
 switches: learn forwarding table using
flooding, learning, MAC addresses

request
 introduction
 LANs
• addressing, ARP
• Ethernet
• switches
• VLANs
Link Layer: 6-74

Virtual LANs (VLANs): motivation
Link Layer: 6-75
Computer
Science EE
Q: what happens as LAN sizes scale, users change point of attachment?
single broadcast domain:
 scaling: all layer-2 broadcast traffic
(ARP, DHCP, unknown MAC) must
cross entire LAN
 efficiency, security, privacy issues

Virtual LANs (VLANs): motivation
Link Layer: 6-76
administrative issues:
 CS user moves office to EE - physically
attached to EE switch, but wants to
remain logically attached to CS
switch
Computer
Science EE
single broadcast domain:
 scaling: all layer-2 broadcast traffic
(ARP, DHCP, unknown MAC) must
cross entire LAN
 efficiency, security, privacy, efficiency
issues
Q: what happens as LAN sizes scale, users change point of attachment?

1
8
2
7 9
16
10
15
Port-based VLANs
Link Layer: 6-77
switch(es) supporting
VLAN capabilities can be
configured to define
multiple virtual LANS
over single physical LAN
infrastructure.
Virtual Local Area
Network (VLAN)
port-based VLAN: switch ports grouped (by
switch management software) so that
single physical switch ……
…
EE (VLAN ports 1-8) CS (VLAN ports 9-15)
…
… operates as multiple virtual switches
1
8
2
7
EE (VLAN ports 1-8)
…
9
16
10
15
…
CS (VLAN ports 9-15)

1
8
2
7 9
16
10
15
Port-based VLANs
Link Layer: 6-78
…
…
 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
switch port
 dynamic membership: ports can be
dynamically assigned among VLANs
 forwarding between VLANS: done via
routing (just as with separate switches)
• in practice vendors sell combined switches
plus routers

1
8
2
7 9
16
10
15
VLANS spanning multiple switches
Link Layer: 6-79
…
…
5
8
2
7
…
16
1
6
3
4
Ports 2,3,5 belong to EE VLAN
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

802.1Q VLAN frame format
Link Layer: 6-80
802.1 Ethernet frame
dest.
address
source
preamble
type
2-byte Tag Protocol Identifier
(value: 81-00) Tag Control Information
(12 bit VLAN ID field, 3 bit priority field like IP
TOS)
Recomputed
CRC
802.1Q frame
dest.
address
source
preamble
type

1
8
2
7 9
16
10
15
EVPN: Ethernet VPNs (aka VXLANs)
Link Layer: 6-81
…
…
Layer-2 Ethernet switches logically connected to each other (e.g., using IP as an
underlay)
 Ethernet frames carried within IP datagrams between sites
 “tunneling scheme to overlay Layer 2 networks on top of Layer 3 networks ... runs over the existing
networking infrastructure and provides a means to "stretch" a Layer 2 network.” [RFC 7348]
5
8
2
7
…
6
3
4
16
1
Sunnyvale
data center
Bangalore
data center
Ethernet
Ethernet
frame
IP
datagram

request
 introduction
 LANs
• addressing, ARP
• Ethernet
• switches
• VLANs
Link Layer: 6-82

label Exp S TTL
20 3 1 5
Multiprotocol label switching (MPLS)
Link Layer: 6-83
 goal: high-speed IP forwarding among network of MPLS-capable
routers, using fixed length label (instead of shortest prefix matching)
• faster lookup using fixed length identifier
• borrowing ideas from Virtual Circuit (VC) approach
• but IP datagram still keeps IP address!
remainder of Ethernet frame, including IP
header with IP source, destination addresses
MPLS header
Ethernet
header
remainder of Ethernet frame, including IP
header with IP source, destination addresses

MPLS capable routers
Link Layer: 6-84
 a.k.a. label-switched router
 forward packets to outgoing interface based only on label
value (don’t inspect IP address)
• MPLS forwarding table distinct from IP forwarding tables
 flexibility: MPLS forwarding decisions can differ from
those of IP
• use destination and source addresses to route flows to same
destination differently (traffic engineering)
• re-route flows quickly if link fails: pre-computed backup paths

MPLS versus IP paths
Link Layer: 6-85
R2
D
R3
R5
A
R6
R4
 IP routing: path to destination determined by destination address alone
IP router

MPLS versus IP paths
Link Layer: 6-86
R2
D
R3
R5
A
R6
IP router
R4
 IP routing: path to destination determined by destination address alone
IP/MPLS router
IP/MPLS entry router (R4) can use different MPLS routes to
A based, e.g., on IP source address or other fields
 MPLS routing: path to destination can be based on source and
destination address
• flavor of generalized forwarding (MPLS 10 years earlier)
• fast reroute: precompute backup routes in case of link failure
R1

MPLS signaling
Link Layer: 6-87
 modify OSPF, IS-IS link-state flooding protocols to carry info
used by MPLS routing:
• e.g., link bandwidth, amount of “reserved” link bandwidth
R2
D
R3
R5
A
R6
R4
modified
link state
flooding
RSVP-TE
 entry MPLS router uses RSVP-TE signaling protocol to set up
MPLS forwarding at downstream routers
R1

MPLS forwarding tables
Link Layer: 6-88
in out out
label label dest
interface
6 - A 0
in out out
label label dest
interface
10 6 A 1
12 9 D 0
in out out
label label dest
interface
8 6 A 0
in out out
label label dest
interface
10 A 0
12 D 0
8 A 1
R2
D
R3
R5
A
R6
R4
R1
0
1
0
0
1
0

request
 introduction
 LANs
• addressing, ARP
• Ethernet
• switches
• VLANs
Link Layer: 6-89

Datacenter networks
Link Layer: 6-90
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
 reliability
 managing/balancing load, avoiding
processing, networking, data
bottlenecks Inside a 40-ft Microsoft container, Chicago data center

Datacenter networks: network elements
Link Layer: 6-91
Server racks
 20- 40 server blades: hosts
Top of Rack (TOR) switch
 one per rack
 100G-400G Ethernet to blades
Tier-2 switches
 connecting to ~16 TORs below
Tier-1 switches
 connecting to ~16 T-2s below
Border routers
 connections outside datacenter
…
…
…
…
…
…
…
…

Datacenter networks: network elements
Link Layer: 6-92
Facebook F16 data center network topology:
https://engineering.fb.com/data-center-engineering/f16-minipack/ (posted 3/2019)

Datacenter networks: multipath
Link Layer: 6-93
9 10 11 12 13 14 15 16
two disjoint paths highlighted between racks 1 and 11
 rich interconnection among switches, racks:
• increased throughput between racks (multiple routing paths possible)
• increased reliability via redundancy

…
…
…
…
…
…
…
…
Datacenter networks: application-layer routing
Link Layer: 6-94
Load
balancer
Internet
load balancer:
application-layer
routing
 receives external
client requests
 directs workload
within data center
 returns results to
external client
(hiding data center
internals from client)

 link layer:
• RoCE: remote DMA (RDMA) over Converged Ethernet
 transport layer:
• ECN (explicit congestion notification) used in transport-layer congestion
control (DCTCP, DCQCN)
• experimentation with hop-by-hop (backpressure) congestion control
 routing, management:
• SDN widely used within/among organizations’ datacenters
• place related services, data as close as possible (e.g., in same rack or nearby
rack) to minimize tier-2, tier-1 communication
Datacenter networks: protocol innovations
Link Layer: 6-95
Google Networking: Infrastructure and Selected Challenges (Slides:
https://networkingchannel.eu/google-networking-infrastructure-and-selected-challenges/

ORION: Google’s new SDN control plane for internal
datacenter (Jupiter) + wide area (B4) network
Orion SDN architecture and core apps
 routing (intradomain, iBGP), traffic
engineering: implemented in applications
on top of ORION core
 edge-edge flow-based controls (e.g.,
CoFlow scheduling) to meet contract SLAs
 management: pub-sub distributed
microservices in Orion core, OpenFlow for
switch signaling/monitoring
Note:
 no routing protocols, congestion control (partially) also managed by SDN rather
than by protocol
 are protocols dying?
Network Layer Control Plane: 5-96

request
 introduction
 LANs
• addressing, ARP
• Ethernet
• switches
• VLANs
Link Layer: 6-97

Synthesis: a day in the life of a web request
Link Layer: 6-98
 our journey down the protocol stack is now 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

A day in the life: scenario
Link Layer: 6-99
Comcast network
68.80.0.0/13
Google’s network
64.233.160.0/19
64.233.169.105
web server
DNS server
school network
68.80.2.0/24
browser
web page
 arriving mobile
client attaches
to network …
 requests web
page:
www.google.com
scenario:
Sounds
simple!

A day in the life: connecting to the Internet
Link Layer: 6-100
router has
DHCP server
arriving mobile:
DHCP client
 connecting laptop needs to get its own IP
address, addr of first-hop router, addr of
DNS server: use DHCP
DHCP
UDP
IP
Eth
Phy
DHCP
DHCP
DHCP
DHCP
DHCP
DHCP
UDP
IP
Eth
Phy
DHCP
DHCP
DHCP
DHCP
DHCP
 DHCP request encapsulated in UDP,
encapsulated in IP, encapsulated in 802.3
Ethernet
 Ethernet frame broadcast (dest:
FFFFFFFFFFFF) on LAN, received at router
running DHCP server
 Ethernet de-muxed to IP de-muxed, UDP
de-muxed to DHCP

A day in the life: connecting to the Internet
Link Layer: 6-101
router has
DHCP server
arriving mobile:
DHCP client
DHCP
UDP
IP
Eth
Phy
DHCP
UDP
IP
Eth
Phy
 DHCP server formulates DHCP ACK
containing client’s IP address, IP address
of first-hop router for client, name & IP
address of DNS server
DHCP
DHCP
DHCP
DHCP
DHCP
DHCP
DHCP
DHCP
DHCP  encapsulation at DHCP server, frame
forwarded (switch learning) through LAN,
demultiplexing at client
Client now has IP address, knows name & addr of DNS
server, IP address of its first-hop router
 DHCP client receives DHCP ACK reply

A day in the life… ARP (before DNS, before HTTP)
Link Layer: 6-102
router has
ARP server
arriving mobile:
ARP client
DNS
UDP
IP
Eth
Phy
Eth
Phy
ARP
 before sending HTTP request, need IP address
of www.google.com: DNS
DNS
DNS
DNS
 DNS query created, encapsulated in UDP,
encapsulated in IP, encapsulated in Eth. To
send frame to router, need MAC address of
router interface: ARP
 ARP query broadcast, received by 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
ARP query
ARP
ARP reply

A day in the life… using DNS
Link Layer: 6-103
DNS
UDP
IP
Eth
Phy
Comcast network
68.80.0.0/13
DNS
server
DNS
DNS
DNS
DNS
DNS
 IP datagram
containing DNS query
forwarded via LAN
switch from client to
1st
hop router
 IP datagram forwarded from campus
network into Comcast network,
routed (tables created by RIP, OSPF,
IS-IS and/or BGP routing protocols)
to DNS server
 de-muxed to DNS
 DNS replies to client
with IP address of
www.google.com
DNS
UDP
IP
Eth
Phy
DNS
DNS
DNS
DNS
DNS

A day in the life…TCP connection carrying HTTP
Link Layer: 6-104
Comcast network
68.80.0.0/13
64.233.169.105
Google web server
HTTP
TCP
IP
Eth
Phy
HTTP
 to send HTTP request,
client first opens TCP
socket to web server
 TCP SYN segment (step 1 in TCP
3-way handshake) inter-domain
routed to web server
 TCP connection established!
SYN
SYN
SYN
SYN
TCP
IP
Eth
Phy
SYN
SYN
SYN
SYNACK
SYNACK
SYNACK
SYNACK
SYNACK
SYNACK
SYNACK
 web server responds with
TCP SYNACK (step 2 in TCP 3-
way handshake)

A day in the life… HTTP request/reply
Link Layer: 6-105
Comcast network
68.80.0.0/13
64.233.169.105
Google web server
HTTP
TCP
IP
Eth
Phy
HTTP
TCP
IP
Eth
Phy
HTTP
 HTTP request sent into
TCP socket
 IP datagram containing
HTTP request routed to
www.google.com
 IP datagram containing
HTTP reply routed back to
client
 web server responds with
HTTP reply (containing web
page)
HTTP
HTTP
HTTP
HTTP
HTTP
HTTP
HTTP
HTTP
HTTP
HTTP
HTTP
HTTP
HTTP
 web page finally (!!!)
displayed

Chapter 6: Summary
Link Layer: 6-106
 principles behind data link layer services:
• error detection, correction
• sharing a broadcast channel: multiple access
• link layer addressing
 instantiation, 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

Chapter 6: let’s take a breath
Link Layer: 6-107
 journey down protocol stack complete (except PHY)
 solid understanding of networking principles, practice!
 ….. could stop here …. but more interesting topics!
• wireless
• security

Additional Chapter 6 slides
Network Layer: 5-108

Pure ALOHA efficiency
Link Layer: 6-109
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!

Machine Learning (ML) is a branch of Artificial Intelligence (AI)

More Related Content

Recently uploaded (20)

Featured (20)

Machine Learning (ML) is a branch of Artificial Intelligence (AI)

Editor's Notes