Chapter 3

Transport Layer

A note on the use of these ppt slides:

Were making these slides freely available to all (faculty, students, readers).
Theyre in PowerPoint form so you can add, modify, and delete slides
(including this one) and slide content to suit your needs. They obviously
Computer Networking:
represent a lot of work on our part. In return for use, we only ask the A Top Down Approach
following:
If you use these slides (e.g., in a class) in substantially unaltered form,
4th edition.
that you mention their source (after all, wed like people to use our book!) Jim Kurose, Keith Ross
If you post any slides in substantially unaltered form on a www site, that
Addison-Wesley, July

al
you note that they are adapted from (or perhaps identical to) our slides, and
note our copyright of this material. 2007.

ep
Thanks and enjoy! JFK/KWR

itn
All material copyright 1996-2007
J.F Kurose and K.W. Ross, All Rights Reserved

cs
Transport Layer 3-1
Chapter 3: Transport Layer
Our goals:
understand principles
behind transport
layer services:
multiplexing/demultipl
exing
reliable data transfer
flow control
congestion control
learn about transport
layer protocols in the
Internet:
UDP: connectionless
transport
TCP: connection-oriented
transport
TCP congestion control

al
ep
itn
cs
Transport Layer 3-2
Chapter 3 outline
3.1 Transport-layer 3.5 Connection-oriented
services transport: TCP
3.2 Multiplexing and segment structure
demultiplexing reliable data transfer
flow control
3.3 Connectionless

connection management
transport: UDP

3.6 Principles of
3.4 Principles of
reliable data transfer congestion control
3.7 TCP congestion
control

al
ep
itn
cs
Transport Layer 3-3
Transport services and protocols
applicatio
n
provide logical communication transport
network
between app processes data link
physical
running on different hosts
transport protocols run in
end systems
send side: breaks app
messages into segments,
passes to network layer
rcv side: reassembles applicatio
n
segments into messages, transport
network
passes to app layer data link
physical

more than one transport

al
protocol available to apps

ep
Internet: TCP and UDP

itn
cs
Transport Layer 3-4
Transport vs. network layer
network layer: logical Household analogy:
communication 12 kids sending letters to
between hosts 12 kids
transport layer: logical processes = kids
communication app messages = letters
between processes in envelopes
relies on, enhances,
hosts = houses

network layer services
transport protocol =
Ann and Bill

al
network-layer protocol

ep
= postal service

itn
cs
Transport Layer 3-5
Internet transport-layer protocols
reliable, in-order
applicatio
n
transport
delivery (TCP) network
data link
network
congestion control physical
data link
network
physical
data link
flow control physical

connection setup
unreliable, unordered
network
data link
physicalnetwork
delivery: UDP data link
physical
no-frills extension of network
data link
best-effort IP physical network
applicatio
n
data link transport
services not available:
physical network
data link

al
physical
delay guarantees

ep
bandwidth guarantees

itn
cs
Transport Layer 3-6
Chapter 3 outline
3.1 Transport-layer 3.5 Connection-oriented
services transport: TCP
3.2 Multiplexing and segment structure
demultiplexing reliable data transfer
flow control
3.3 Connectionless

connection management
transport: UDP

3.6 Principles of
3.4 Principles of
reliable data transfer congestion control
3.7 TCP congestion
control

al
ep
itn
cs
Transport Layer 3-7
 Multiplexing/demultiplexing
Demultiplexing at rcv host: Multiplexing at send host:
gathering data from multiple
delivering received segments
sockets, enveloping data with
to correct socket
header (later used for
demultiplexing)
= socket = process

P3 P1
P1 P2 P4 application
application application

transport transport transport

network network network

link link link

al
physical

ep
physical physical
host 3

itn
host 1 host 2

cs
Transport Layer 3-8
How demultiplexing works
host receives IP datagrams
each datagram has source 32 bits
IP address, destination IP
address source port # dest port #

each datagram carries 1

transport-layer segment other header fields
each segment has source,
destination port number
host uses IP addresses & port application
numbers to direct segment to data
appropriate socket (message)

al
ep
TCP/UDP segment format

itn
cs
Transport Layer 3-9
Connectionless demultiplexing
Create sockets with port
numbers:
DatagramSocket mySocket1 = new
DatagramSocket(12534);
DatagramSocket mySocket2 = new
DatagramSocket(12535);
UDP socket identified by
two-tuple:
(dest IP address, dest port number)
When host receives UDP
segment:
checks destination port
number in segment
directs UDP segment to
socket with that port
number
IP datagrams with
different source IP
addresses and/or source
port numbers directed

al
to same socket

ep
itn
cs
Transport Layer 3-10
 Connectionless demux (cont)
DatagramSocket serverSocket = new DatagramSocket(6428);

P2 P1
P1
P3

SP: 6428 SP: 6428

DP: 9157 DP: 5775

SP: 9157 SP: 5775

client DP: 6428 DP: 6428 Client
server
IP: A IP: C IP:B

al
ep
SP provides return address

itn
cs
Transport Layer 3-11
 Connection-oriented demux
TCP socket identified Server host may support
by 4-tuple: many simultaneous TCP
source IP address sockets:
source port number each socket identified by
dest IP address its own 4-tuple
dest port number Web servers have
recv host uses all four different sockets for
values to direct each connecting client
segment to appropriate non-persistent HTTP will
socket have different socket for

al
each request

ep
itn
cs
Transport Layer 3-12
Connection-oriented demux
(cont)

P1 P4 P5 P6 P2 P1P3

SP: 5775
DP: 80
S-IP: B
D-IP:C

SP: 9157 SP: 9157

client DP: 80 DP: 80 Client
server
IP: A S-IP: A
IP: C S-IP: B IP:B

al
D-IP:C D-IP:C

ep
itn
cs
Transport Layer 3-13
Connection-oriented demux:
Threaded Web Server

P1 P4 P2 P1P3

SP: 5775
DP: 80
S-IP: B
D-IP:C

SP: 9157 SP: 9157

client DP: 80 DP: 80 Client
server
IP: A S-IP: A
IP: C S-IP: B IP:B

al
D-IP:C D-IP:C

ep
itn
cs
Transport Layer 3-14
Chapter 3 outline
3.1 Transport-layer 3.5 Connection-oriented
services transport: TCP
3.2 Multiplexing and segment structure
demultiplexing reliable data transfer
flow control
3.3 Connectionless

connection management
transport: UDP

3.6 Principles of
3.4 Principles of
reliable data transfer congestion control
3.7 TCP congestion
control

al
ep
itn
cs
Transport Layer 3-15
UDP: User Datagram Protocol [RFC 768]
no frills, bare bones
Internet transport Why is there a UDP?
protocol
no connection
best effort service, UDP establishment (which can
segments may be: add delay)
lost simple: no connection state
delivered out of order at sender, receiver
to app small segment header
connectionless: no congestion control: UDP
no handshaking between can blast away as fast as
UDP sender, receiver desired
each UDP segment

al

handled independently

ep
of others

itn
cs
Transport Layer 3-16
UDP: more
often used for streaming
multimedia apps 32 bits

loss tolerant Length, in source port # dest port #

rate sensitive bytes of UDP length checksum
segment,
other UDP uses including
DNS header
SNMP
reliable transfer over UDP: Application
add reliability at data
application layer (message)
application-specific

al
error recovery!

ep
UDP segment format

itn
cs
Transport Layer 3-17
UDP checksum
Goal: detect errors (e.g., flipped bits) in transmitted
segment

Sender: Receiver:
treat segment contents compute checksum of
as sequence of 16-bit received segment
integers check if computed checksum
checksum: addition (1s equals checksum field value:
complement sum) of NO - error detected
segment contents YES - no error detected.
sender puts checksum But maybe errors

al
value into UDP checksum nonetheless? More later

ep
field .

itn
cs
Transport Layer 3-18
 Internet Checksum Example
Note
When adding numbers, a carryout from the
most significant bit needs to be added to the
result
Example: add two 16-bit integers

1 1 1 1 0 0 1 1 0 0 1 1 0 0 1 1 0
1 1 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1

wraparound 1 1 0 1 1 1 0 1 1 1 0 1 1 1 0 1 1

al
ep
sum 1 1 0 1 1 1 0 1 1 1 0 1 1 1 1 0 0
checksum 1 0 1 0 0 0 1 0 0 0 1 0 0 0 0 1 1

itn
cs
Transport Layer 3-19
Chapter 3 outline
3.1 Transport-layer 3.5 Connection-oriented
services transport: TCP
3.2 Multiplexing and segment structure
demultiplexing reliable data transfer
flow control
3.3 Connectionless

connection management
transport: UDP

3.6 Principles of
3.4 Principles of
reliable data transfer congestion control
3.7 TCP congestion
control

al
ep
itn
cs
Transport Layer 3-20
Principles of Reliable data transfer
important in app., transport, link layers
top-10 list of important networking topics!

al
ep
characteristics of unreliable channel will determine
complexity of reliable data transfer protocol (rdt)

itn
cs
Transport Layer 3-21
Principles of Reliable data transfer
important in app., transport, link layers
top-10 list of important networking topics!

al
ep
characteristics of unreliable channel will determine
complexity of reliable data transfer protocol (rdt)

itn
cs
Transport Layer 3-22
Principles of Reliable data transfer
important in app., transport, link layers
top-10 list of important networking topics!

al
ep
characteristics of unreliable channel will determine
complexity of reliable data transfer protocol (rdt)

itn
cs
Transport Layer 3-23
 Reliable data transfer: getting started
rdt_send(): called from above, deliver_data(): called by
(e.g., by app.). Passed data to rdt to deliver data to upper
deliver to receiver upper layer

send receive
side side

al
udt_send(): called by rdt, rdt_rcv(): called when packet

ep
to transfer packet over arrives on rcv-side of channel
unreliable channel to receiver

itn
cs
Transport Layer 3-24
Reliable data transfer: getting started
Well:
incrementally develop sender, receiver sides of
reliable data transfer protocol (rdt)
consider only unidirectional data transfer
but control info will flow on both directions!
use finite state machines (FSM) to specify
sender, receiver
event causing state transition
actions taken on state transition
state: when in this
state next state state state
1 event

al
uniquely determined 2
actions

ep
by next event

itn
cs
Transport Layer 3-25
Rdt1.0: reliable transfer over a reliable channel
underlying channel perfectly reliable
no bit errors
no loss of packets

separate FSMs for sender, receiver:

sender sends data into underlying channel
receiver read data from underlying channel

Wait for rdt_send(data) Wait for rdt_rcv(packet)

call from call from extract (packet,data)
above packet = make_pkt(data) below deliver_data(data)
udt_send(packet)

al
ep
sender receiver

itn
cs
Transport Layer 3-26
Rdt2.0: channel with bit errors
underlying channel may flip bits in packet
checksum to detect bit errors

the question: how to recover from errors:

acknowledgements (ACKs): receiver explicitly tells sender
that pkt received OK
negative acknowledgements (NAKs): receiver explicitly
tells sender that pkt had errors
sender retransmits pkt on receipt of NAK
new mechanisms in rdt2.0 (beyond rdt1.0):
error detection
receiver feedback: control msgs (ACK,NAK) rcvr->sender

al

ep
itn
cs
Transport Layer 3-27
rdt2.0: FSM specification
rdt_send(data)
snkpkt = make_pkt(data, checksum) receiver
udt_send(sndpkt)
rdt_rcv(rcvpkt) &&
isNAK(rcvpkt)
Wait for Wait for rdt_rcv(rcvpkt) &&
call from ACK or udt_send(sndpkt) corrupt(rcvpkt)
above NAK
udt_send(NAK)

rdt_rcv(rcvpkt) && isACK(rcvpkt)

Wait for
L
call from
sender below

rdt_rcv(rcvpkt) &&
notcorrupt(rcvpkt)

al
extract(rcvpkt,data)

ep
deliver_data(data)
udt_send(ACK)

itn
cs
Transport Layer 3-28
rdt2.0: operation with no errors
rdt_send(data)
snkpkt = make_pkt(data, checksum)
udt_send(sndpkt)
rdt_rcv(rcvpkt) &&
isNAK(rcvpkt)
Wait for Wait for rdt_rcv(rcvpkt) &&
call from ACK or udt_send(sndpkt) corrupt(rcvpkt)
above NAK
udt_send(NAK)

rdt_rcv(rcvpkt) && isACK(rcvpkt)

Wait for
L call from
below

rdt_rcv(rcvpkt) &&
notcorrupt(rcvpkt)

al
extract(rcvpkt,data)

ep
deliver_data(data)
udt_send(ACK)

itn
cs
Transport Layer 3-29
rdt2.0: error scenario
rdt_send(data)
snkpkt = make_pkt(data, checksum)
udt_send(sndpkt)
rdt_rcv(rcvpkt) &&
isNAK(rcvpkt)
Wait for Wait for rdt_rcv(rcvpkt) &&
call from ACK or udt_send(sndpkt) corrupt(rcvpkt)
above NAK
udt_send(NAK)

rdt_rcv(rcvpkt) && isACK(rcvpkt)

Wait for
L call from
below

rdt_rcv(rcvpkt) &&
notcorrupt(rcvpkt)

al
extract(rcvpkt,data)

ep
deliver_data(data)
udt_send(ACK)

itn
cs
Transport Layer 3-30
rdt2.0 has a fatal flaw!
What happens if Handling duplicates:
ACK/NAK corrupted? sender retransmits current
sender doesnt know what pkt if ACK/NAK garbled
happened at receiver! sender adds sequence
cant just retransmit: number to each pkt
possible duplicate receiver discards (doesnt
deliver up) duplicate pkt

stop and wait

Sender sends one packet,
then waits for receiver

al
response

ep
itn
cs
Transport Layer 3-31
rdt2.1: sender, handles garbled ACK/NAKs
rdt_send(data)
sndpkt = make_pkt(0, data, checksum)
udt_send(sndpkt) rdt_rcv(rcvpkt) &&
( corrupt(rcvpkt) ||
Wait for Wait for
ACK or
isNAK(rcvpkt) )
call 0 from
NAK 0 udt_send(sndpkt)
above
rdt_rcv(rcvpkt)
&& notcorrupt(rcvpkt) rdt_rcv(rcvpkt)
&& isACK(rcvpkt) && notcorrupt(rcvpkt)
&& isACK(rcvpkt)
L
L
Wait for Wait for
ACK or call 1 from
rdt_rcv(rcvpkt) && NAK 1 above
( corrupt(rcvpkt) ||
rdt_send(data)

al
isNAK(rcvpkt) )
sndpkt = make_pkt(1, data, checksum)

ep
udt_send(sndpkt)
udt_send(sndpkt)

itn
cs
Transport Layer 3-32
 rdt2.1: receiver, handles garbled ACK/NAKs
rdt_rcv(rcvpkt) && notcorrupt(rcvpkt)
&& has_seq0(rcvpkt)
extract(rcvpkt,data)
deliver_data(data)
sndpkt = make_pkt(ACK, chksum)
udt_send(sndpkt)
rdt_rcv(rcvpkt) && (corrupt(rcvpkt) rdt_rcv(rcvpkt) && (corrupt(rcvpkt)
sndpkt = make_pkt(NAK, chksum) sndpkt = make_pkt(NAK, chksum)
udt_send(sndpkt) udt_send(sndpkt)
Wait for Wait for
rdt_rcv(rcvpkt) && 0 from 1 from rdt_rcv(rcvpkt) &&
not corrupt(rcvpkt) && below below not corrupt(rcvpkt) &&
has_seq1(rcvpkt) has_seq0(rcvpkt)
sndpkt = make_pkt(ACK, chksum) sndpkt = make_pkt(ACK, chksum)
udt_send(sndpkt) udt_send(sndpkt)
rdt_rcv(rcvpkt) && notcorrupt(rcvpkt)
&& has_seq1(rcvpkt)

al
extract(rcvpkt,data)

ep
deliver_data(data)
sndpkt = make_pkt(ACK, chksum)

itn
udt_send(sndpkt)

cs
Transport Layer 3-33
rdt2.1: discussion
Sender:
seq # added to pkt
two seq. #s (0,1) will
suffice. Why?
must check if received
ACK/NAK corrupted
twice as many states
state must remember ACK/NAK received OK
whether current pkt
has 0 or 1 seq. #
Receiver:
must check if received
packet is duplicate
state indicates whether
0 or 1 is expected pkt
seq #
note: receiver can not
know if its last
at sender

al
ep
itn
cs
Transport Layer 3-34
rdt2.2: a NAK-free protocol

same functionality as rdt2.1, using ACKs only

instead of NAK, receiver sends ACK for last pkt
received OK
receiver must explicitly include seq # of pkt being ACKed
duplicate ACK at sender results in same action as
NAK: retransmit current pkt

al
ep
itn
cs
Transport Layer 3-35
 rdt2.2: sender, receiver fragments
rdt_send(data)
sndpkt = make_pkt(0, data, checksum)
udt_send(sndpkt)
rdt_rcv(rcvpkt) &&
( corrupt(rcvpkt) ||
Wait for Wait for
ACK isACK(rcvpkt,1) )
call 0 from
above 0 udt_send(sndpkt)
sender FSM
fragment rdt_rcv(rcvpkt)
&& notcorrupt(rcvpkt)
rdt_rcv(rcvpkt) && && isACK(rcvpkt,0)
(corrupt(rcvpkt) || L
has_seq1(rcvpkt)) Wait for receiver FSM
0 from
udt_send(sndpkt) below fragment

al
rdt_rcv(rcvpkt) && notcorrupt(rcvpkt)
&& has_seq1(rcvpkt)

ep
extract(rcvpkt,data)

itn
deliver_data(data)
sndpkt = make_pkt(ACK1, chksum)

cs
udt_send(sndpkt) Transport Layer 3-36
rdt3.0: channels with errors and loss
New assumption:
underlying channel can
also lose packets (data
or ACKs)
checksum, seq. #, ACKs,
retransmissions will be
of help, but not enough
Approach: sender waits
reasonable amount of
time for ACK
retransmits if no ACK
received in this time
if pkt (or ACK) just delayed
(not lost):
retransmission will be
duplicate, but use of seq.
#s already handles this
receiver must specify seq
al
# of pkt being ACKed
ep
requires countdown timer
itn
cs
Transport Layer 3-37
rdt3.0 sender
rdt_send(data)
rdt_rcv(rcvpkt) &&
sndpkt = make_pkt(0, data, checksum) ( corrupt(rcvpkt) ||
udt_send(sndpkt) isACK(rcvpkt,1) )
rdt_rcv(rcvpkt) start_timer L
L Wait for Wait
for timeout
call 0from
ACK0 udt_send(sndpkt)
above
start_timer
rdt_rcv(rcvpkt)
&& notcorrupt(rcvpkt) rdt_rcv(rcvpkt)
&& isACK(rcvpkt,1) && notcorrupt(rcvpkt)
stop_timer && isACK(rcvpkt,0)
stop_timer
Wait Wait for
timeout for call 1 from
udt_send(sndpkt) ACK1 above
start_timer rdt_rcv(rcvpkt)
rdt_send(data) L
rdt_rcv(rcvpkt) &&

al
( corrupt(rcvpkt) || sndpkt = make_pkt(1, data, checksum)

ep
isACK(rcvpkt,0) ) udt_send(sndpkt)
start_timer
L

itn
cs
Transport Layer 3-38
rdt3.0 in action

al
ep
itn
cs
Transport Layer 3-39
rdt3.0 in action

al
ep
itn
cs
Transport Layer 3-40
Performance of rdt3.0

rdt3.0 works, but performance stinks

ex: 1 Gbps link, 15 ms prop. delay, 8000 bit packet:

L 8000bits
dtrans 9
8 microsecon ds
R 10 bps
U sender: utilization fraction of time sender busy sending

U L/R .008
sender
= = = 0.00027
RTT + L / R 30.008 microsec
onds
1KB pkt every 30 msec -> 33kB/sec thruput over 1 Gbps link

al

network protocol limits use of physical resources!

ep

itn
cs
Transport Layer 3-41
 rdt3.0: stop-and-wait operation
sender receiver
first packet bit transmitted, t = 0
last packet bit transmitted, t = L / R

first packet bit arrives

RTT last packet bit arrives, send ACK

ACK arrives, send next

packet, t = RTT + L / R

L/R .008

al
U = = = 0.00027
sender

ep
RTT + L / R 30.008 microsec
onds

itn
cs
Transport Layer 3-42
Pipelined protocols
Pipelining: sender allows multiple, in-flight, yet-to-
be-acknowledged pkts
range of sequence numbers must be increased
buffering at sender and/or receiver

al
Two generic forms of pipelined protocols: go-Back-N,

ep
selective repeat

itn
cs
Transport Layer 3-43
 Pipelining: increased utilization
sender receiver
first packet bit transmitted, t = 0
last bit transmitted, t = L / R

first packet bit arrives

RTT last packet bit arrives, send ACK
last bit of 2nd packet arrives, send ACK
last bit of 3rd packet arrives, send ACK
ACK arrives, send next
packet, t = RTT + L / R

Increase utilization
by a factor of 3!

al
U 3*L/R .024
= = = 0.0008

ep
sender 30.008
RTT + L / R microsecon

itn
ds

cs
Transport Layer 3-44
Pipelining Protocols
Go-back-N: big picture:
Sender can have up to
N unacked packets in
pipeline
Rcvr only sends
cumulative acks
Doesnt ack packet if Sender maintains
theres a gap
Sender has timer for
oldest unacked packet
If timer expires,
Selective Repeat: big pic
Sender can have up to
N unacked packets in
pipeline
Rcvr acks individual
packets
timer for each
unacked packet
When timer expires,
retransmit only unack

al

retransmit all unacked packet

ep
packets

itn
cs
Transport Layer 3-45
Selective repeat: big picture
Sender can have up to N unacked packets
in pipeline
Rcvr acks individual packets
Sender maintains timer for each unacked
packet
When timer expires, retransmit only unack
packet

al
ep
itn
cs
Transport Layer 3-46
Go-Back-N
Sender:
k-bit seq # in pkt header
window of up to N, consecutive unacked pkts allowed

ACK(n): ACKs all pkts up to, including seq # n - cumulative ACK

may receive duplicate ACKs (see receiver)

al
timer for each in-flight pkt

ep
timeout(n): retransmit pkt n and all higher seq # pkts in window

itn
cs
Transport Layer 3-47
GBN: sender extended FSM
rdt_send(data)
if (nextseqnum < base+N) {
sndpkt[nextseqnum] = make_pkt(nextseqnum,data,chksum)
udt_send(sndpkt[nextseqnum])
if (base == nextseqnum)
start_timer
nextseqnum++
}
L else
refuse_data(data)
base=1
nextseqnum=1
timeout
start_timer
Wait
udt_send(sndpkt[base])
rdt_rcv(rcvpkt) udt_send(sndpkt[base+1])
&& corrupt(rcvpkt)
udt_send(sndpkt[nextseqnum-1])
rdt_rcv(rcvpkt) &&

al
notcorrupt(rcvpkt)

ep
base = getacknum(rcvpkt)+1
If (base == nextseqnum)

itn
stop_timer
else

cs
start_timer Transport Layer 3-48
GBN: receiver extended FSM
default
udt_send(sndpkt) rdt_rcv(rcvpkt)
&& notcurrupt(rcvpkt)
L && hasseqnum(rcvpkt,expectedseqnum)
expectedseqnum=1 Wait extract(rcvpkt,data)
sndpkt = deliver_data(data)
make_pkt(expectedseqnum,ACK,chksum) sndpkt = make_pkt(expectedseqnum,ACK,chksum)
udt_send(sndpkt)
expectedseqnum++

ACK-only: always send ACK for correctly-received pkt

with highest in-order seq #
may generate duplicate ACKs
need only remember expectedseqnum
out-of-order pkt:

al
ep
discard (dont buffer) -> no receiver buffering!

itn
Re-ACK pkt with highest in-order seq #

cs
Transport Layer 3-49
GBN in
action

al
ep
itn
cs
Transport Layer 3-50
Selective Repeat
receiver individually acknowledges all correctly
received pkts
buffers pkts, as needed, for eventual in-order delivery
to upper layer
sender only resends pkts for which ACK not
received
sender timer for each unACKed pkt
sender window
N consecutive seq #s
again limits seq #s of sent, unACKed pkts

al
ep
itn
cs
Transport Layer 3-51
Selective repeat: sender, receiver windows

al
ep
itn
cs
Transport Layer 3-52
Selective repeat
sender
data from above :
if next available seq # in
window, send pkt
timeout(n):
resend pkt n, restart timer
ACK(n) in [sendbase,sendbase+N]:
mark pkt n as received
if n smallest unACKed pkt,
advance window base to
next unACKed seq #
receiver
pkt n in [rcvbase, rcvbase+N-1]
send ACK(n)
out-of-order: buffer
in-order: deliver (also
deliver buffered, in-order
pkts), advance window to
next not-yet-received pkt
pkt n in [rcvbase-N,rcvbase-1]
ACK(n)
otherwise:

al
ignore

ep
itn
cs
Transport Layer 3-53
Selective repeat in action

al
ep
itn
cs
Transport Layer 3-54
Selective repeat:
dilemma
Example:
seq #s: 0, 1, 2, 3
window size=3

receiver sees no
difference in two
scenarios!
incorrectly passes
duplicate data as new
in (a)

al
Q: what relationship

ep
between seq # size

itn
and window size?

cs
Transport Layer 3-55
Chapter 3 outline
3.1 Transport-layer 3.5 Connection-oriented
services transport: TCP
3.2 Multiplexing and segment structure
demultiplexing reliable data transfer
flow control
3.3 Connectionless

connection management
transport: UDP

3.6 Principles of
3.4 Principles of
reliable data transfer congestion control
3.7 TCP congestion
control

al
ep
itn
cs
Transport Layer 3-56
 TCP: Overview RFCs: 793, 1122, 1323, 2018, 2581

point-to-point: full duplex data:

one sender, one receiver bi-directional data flow

reliable, in-order byte in same connection

MSS: maximum segment
steam:
size
no message boundaries
connection-oriented:
pipelined:
handshaking (exchange
TCP congestion and flow
of control msgs) inits
control set window size sender, receiver state
send & receive buffers before data exchange
flow controlled:

al
sender will not

ep
application application
writes data reads data
socket socket

overwhelm receiver
door door

itn
TCP TCP
send buffer receive buffer
segment

cs
Transport Layer 3-57
 TCP segment structure
32 bits
URG: urgent data counting
(generally not used) source port # dest port #
by bytes
sequence number of data
ACK: ACK #
valid acknowledgement number (not segments!)
head not
PSH: push data now len used
UA P R S F Receive window
(generally not used) # bytes
checksum Urg data pnter
rcvr willing
RST, SYN, FIN: to accept
Options (variable length)
connection estab
(setup, teardown
commands)
application
Internet data

al
checksum (variable length)

ep
(as in UDP)

itn
cs
Transport Layer 3-58
 TCP seq. #s and ACKs
Seq. #s:
Host A Host B
byte stream
number of first User
types
byte in segments C
data host ACKs
receipt of
ACKs: C, echoes
seq # of next byte back C
expected from
other side host ACKs
cumulative ACK receipt
of echoed
Q: how receiver handles C
out-of-order segments

al
A: TCP spec doesnt

ep
time
say, - up to
simple telnet scenario
implementor

itn
cs
Transport Layer 3-59
TCP Round Trip Time and Timeout
Q: how to set TCP Q: how to estimate RTT?
timeout value? SampleRTT: measured time from
longer than RTT segment transmission until ACK
but RTT varies
receipt
ignore retransmissions
too short: premature
timeout SampleRTT will vary, want
unnecessary
estimated RTT smoother
retransmissions average several recent

too long: slow reaction

measurements, not just
to segment loss current SampleRTT

al
ep
itn
cs
Transport Layer 3-60
TCP Round Trip Time and Timeout
EstimatedRTT = (1- )*EstimatedRTT + *SampleRTT

Exponential weighted moving average

influence of past sample decreases exponentially fast
typical value: = 0.125

al
ep
itn
cs
Transport Layer 3-61
Example RTT estimation:
RTT: gaia.cs.umass.edu to fantasia.eurecom.fr

350

300

250
RTT (milliseconds)

200

150

al
100

ep
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)

itn
SampleRTT Estimated RTT

cs
Transport Layer 3-62
TCP Round Trip Time and Timeout
Setting the timeout
EstimtedRTT plus safety margin
large variation in EstimatedRTT -> larger safety margin
first estimate of how much SampleRTT deviates from
EstimatedRTT:

DevRTT = (1-)*DevRTT +
*|SampleRTT-EstimatedRTT|

(typically, = 0.25)

Then set timeout interval:

al
ep
TimeoutInterval = EstimatedRTT + 4*DevRTT

itn
cs
Transport Layer 3-63
Chapter 3 outline
3.1 Transport-layer 3.5 Connection-oriented
services transport: TCP
3.2 Multiplexing and segment structure
demultiplexing reliable data transfer
flow control
3.3 Connectionless

connection management
transport: UDP

3.6 Principles of
3.4 Principles of
reliable data transfer congestion control
3.7 TCP congestion
control

al
ep
itn
cs
Transport Layer 3-64
TCP reliable data transfer
TCP creates rdt Retransmissions are
service on top of IPs triggered by:
unreliable service timeout events
Pipelined segments duplicate acks
Cumulative acks Initially consider
TCP uses single
simplified TCP sender:
ignore duplicate acks
retransmission timer
ignore flow control,
congestion control

al
ep
itn
cs
Transport Layer 3-65
TCP sender events:
data rcvd from app: timeout:
Create segment with retransmit segment
seq # that caused timeout
seq # is byte-stream restart timer
number of first data Ack rcvd:
byte in segment If acknowledges
start timer if not previously unacked
already running (think segments
of timer as for oldest update what is known to
unacked segment) be acked
expiration interval: start timer if there are

al
TimeOutInterval outstanding segments

ep
itn
cs
Transport Layer 3-66
NextSeqNum = InitialSeqNum
SendBase = InitialSeqNum

loop (forever) { TCP

sender
switch(event)

event: data received from application above

create TCP segment with sequence number NextSeqNum (simplified)
if (timer currently not running)
start timer
pass segment to IP Comment:
NextSeqNum = NextSeqNum + length(data)
SendBase-1: last
event: timer timeout cumulatively
retransmit not-yet-acknowledged segment with acked byte
smallest sequence number Example:
start timer SendBase-1 = 71;
y= 73, so the rcvr
event: ACK received, with ACK field value of y wants 73+ ;
if (y > SendBase) { y > SendBase, so
SendBase = y
that new data is

al
if (there are currently not-yet-acknowledged segments)
acked

ep
start timer
}

itn
} /* end of loop forever */

cs
Transport Layer 3-67
 TCP: retransmission scenarios
Host A Host B Host A Host B

Seq=92 timeout
timeout

X
loss

Sendbase
= 100

Seq=92 timeout
SendBase
= 120

al
SendBase
SendBase

ep
= 100
= 120 premature timeout

itn
time time
lost ACK scenario

cs
Transport Layer 3-68
 TCP retransmission scenarios (more)
Host A Host B
timeout

X
loss

SendBase
= 120

al
time

ep
Cumulative ACK scenario

itn
cs
Transport Layer 3-69
TCP ACK generation [RFC 1122, RFC 2581]

Event at Receiver TCP Receiver action

Arrival of in-order segment with Delayed ACK. Wait up to 500ms
expected seq #. All data up to for next segment. If no next segment,
expected seq # already ACKed send ACK

Arrival of in-order segment with Immediately send single cumulative

expected seq #. One other ACK, ACKing both in-order segments
segment has ACK pending

Arrival of out-of-order segment Immediately send duplicate ACK,

higher-than-expect seq. # . indicating seq. # of next expected byte
Gap detected

al
ep
Arrival of segment that Immediate send ACK, provided that
partially or completely fills gap segment starts at lower end of gap

itn
cs
Transport Layer 3-70
Fast Retransmit
Time-out period often If sender receives 3
relatively long: ACKs for the same
long delay before data, it supposes that
resending lost packet segment after ACKed
Detect lost segments data was lost:
via duplicate ACKs. fast retransmit: resend
Sender often sends segment before timer
many segments back-to- expires
back
If segment is lost,
there will likely be many

al
duplicate ACKs.

ep
itn
cs
Transport Layer 3-71
 Host A Host B

X
timeout

al
ep
time

itn
cs
Figure 3.37 Resending a segment after triple duplicate ACK Layer
Transport 3-72
 Fast retransmit algorithm:

event: ACK received, with ACK field value of y

if (y > SendBase) {
SendBase = y
if (there are currently not-yet-acknowledged segments)
start timer
}
else {
increment count of dup ACKs received for y
if (count of dup ACKs received for y = 3) {
resend segment with sequence number y
}

al
a duplicate ACK for fast retransmit

ep
already ACKed segment

itn
cs
Transport Layer 3-73
Chapter 3 outline
3.1 Transport-layer 3.5 Connection-oriented
services transport: TCP
3.2 Multiplexing and segment structure
demultiplexing reliable data transfer
flow control
3.3 Connectionless

connection management
transport: UDP

3.6 Principles of
3.4 Principles of
reliable data transfer congestion control
3.7 TCP congestion
control

al
ep
itn
cs
Transport Layer 3-74
TCP Flow Control
flow control
sender wont overflow
receive side of TCP receivers buffer by
connection has a transmitting too
receive buffer: much,
too fast

speed-matching
service: matching the
send rate to the
receiving apps drain
rate
app process may be

al
slow at reading from

ep
buffer

itn
cs
Transport Layer 3-75
TCP Flow control: how it works
Rcvr advertises spare
room by including value
of RcvWindow in
segments
Sender limits unACKed
(Suppose TCP receiver data to RcvWindow
discards out-of-order guarantees receive
segments) buffer doesnt overflow
spare room in buffer
= RcvWindow

al
= RcvBuffer-[LastByteRcvd -

ep
LastByteRead]

itn
cs
Transport Layer 3-76
Chapter 3 outline
3.1 Transport-layer 3.5 Connection-oriented
services transport: TCP
3.2 Multiplexing and segment structure
demultiplexing reliable data transfer
flow control
3.3 Connectionless

connection management
transport: UDP

3.6 Principles of
3.4 Principles of
reliable data transfer congestion control
3.7 TCP congestion
control

al
ep
itn
cs
Transport Layer 3-77
TCP Connection Management
Recall: TCP sender, receiver
establish connection
before exchanging data
segments
initialize TCP variables:
seq. #s
buffers, flow control
info (e.g. RcvWindow)
client: connection initiator
Socket clientSocket = new
Socket("hostname","port
number");
server: contacted by client
Socket connectionSocket =
welcomeSocket.accept();
Three way handshake:
Step 1: client host sends TCP
SYN segment to server
specifies initial seq #
no data
Step 2: server host receives
SYN, replies with SYNACK
segment
server allocates buffers
specifies server initial
seq. #
Step 3: client receives SYNACK,
al
replies with ACK segment,
ep
which may contain data
itn
cs
Transport Layer 3-78
TCP Connection Management (cont.)

Closing a connection: client server

close
client closes socket:
clientSocket.close();

Step 1: client end system close

sends TCP FIN control
segment to server

Step 2: server receives

FIN, replies with ACK. timed wait
Closes connection, sends

al
FIN.

ep
closed

itn
cs
Transport Layer 3-79
TCP Connection Management (cont.)

Step 3: client receives FIN, client server

replies with ACK. closing
Enters timed wait -
will respond with ACK
to received FINs
closing
Step 4: server, receives
ACK. Connection closed.

timed wait
Note: with small
closed
modification, can handle

al
simultaneous FINs.

ep
closed

itn
cs
Transport Layer 3-80
TCP Connection Management (cont)

TCP server
lifecycle

TCP client
lifecycle

al
ep
itn
cs
Transport Layer 3-81
Chapter 3 outline
3.1 Transport-layer 3.5 Connection-oriented
services transport: TCP
3.2 Multiplexing and segment structure
demultiplexing reliable data transfer
flow control
3.3 Connectionless

connection management
transport: UDP

3.6 Principles of
3.4 Principles of
reliable data transfer congestion control
3.7 TCP congestion
control

al
ep
itn
cs
Transport Layer 3-82
Principles of Congestion Control

Congestion:
informally: too many sources sending too much
data too fast for network to handle
different from flow control!
manifestations:
lost packets (buffer overflow at routers)
long delays (queueing in router buffers)
a top-10 problem!

al
ep
itn
cs
Transport Layer 3-83
 Causes/costs of congestion: scenario 1
Host A lout
two senders, two
lin : original data

receivers
one router,
Host B unlimited shared
output link buffers

infinite buffers
no retransmission

large delays
when congested
maximum

al
achievable

ep
throughput

itn
cs
Transport Layer 3-84
 Causes/costs of congestion: scenario 2

one router, finite buffers

sender retransmission of lost packet

Host A lin : original lout

data
l'in : original data, plus
retransmitted data

Host B finite shared output

link buffers

al
ep
itn
cs
Transport Layer 3-85
 Causes/costs of congestion: scenario 2
always: = l
l (goodput)
in out
perfect retransmission only when loss: l > lout
in
retransmission of delayed (not lost) packet makes l larger
in
(than perfect case) for same lout
R/2 R/2 R/2

R/3
lou

lou

lou
R/4
t

t
R/2 R/2 R/2
lin lin lin

a. b. c.

al
costs of congestion:

ep
more work (retrans) for given goodput

itn
unneeded retransmissions: link carries multiple copies of pkt

cs
Transport Layer 3-86
Causes/costs of congestion: scenario 3
four senders
Q: what happens as l
multihop paths in
and l increase ?
timeout/retransmit in
Host A lout
lin : original data
l'in : original data, plus
retransmitted data

finite shared
output link buffers

Host B

al
ep
itn
cs
Transport Layer 3-87
Causes/costs of congestion: scenario 3
H l
o
o
s
u
t
A t

H
o
s
t
B

Another cost of congestion:

when packet dropped, any upstream transmission

al
capacity used for that packet was wasted!

ep
itn
cs
Transport Layer 3-88
Approaches towards congestion control
Two broad approaches towards congestion control:

End-end congestion Network-assisted

control: congestion control:
no explicit feedback from routers provide feedback
network to end systems
congestion inferred from single bit indicating
end-system observed loss, congestion (SNA,
delay DECbit, TCP/IP ECN,
approach taken by TCP ATM)
explicit rate sender

al
should send at

ep
itn
cs
Transport Layer 3-89
Case study: ATM ABR congestion control

ABR: available bit rate: RM (resource management)

elastic service cells:
if senders path sent by sender, interspersed
underloaded: with data cells
sender should use bits in RM cell set by switches
available bandwidth (network-assisted)
if senders path NI bit: no increase in rate
congested: (mild congestion)
sender throttled to CI bit: congestion
minimum guaranteed indication
rate RM cells returned to sender by

al
receiver, with bits intact

ep
itn
cs
Transport Layer 3-90
Case study: ATM ABR congestion control

two-byte ER (explicit rate) field in RM cell

congested switch may lower ER value in cell
sender send rate thus maximum supportable rate on path

EFCI bit in data cells: set to 1 in congested switch

al
if data cell preceding RM cell has EFCI set, sender sets CI

ep
bit in returned RM cell

itn
cs
Transport Layer 3-91
Chapter 3 outline
3.1 Transport-layer 3.5 Connection-oriented
services transport: TCP
3.2 Multiplexing and segment structure
demultiplexing reliable data transfer
flow control
3.3 Connectionless

connection management
transport: UDP

3.6 Principles of
3.4 Principles of
reliable data transfer congestion control
3.7 TCP congestion
control

al
ep
itn
cs
Transport Layer 3-92
 TCP congestion control: additive increase,
multiplicative decrease
Approach: increase transmission rate (window size),
probing for usable bandwidth, until loss occurs
additive increase: increase CongWin by 1 MSS
every RTT until loss detected
multiplicative decrease: cut CongWin in half after
loss congestion
window
congestion window size

24 Kbytes

Saw tooth
behavior: probing
16 Kbytes

for bandwidth

al
8 Kbytes

ep
time

itn
time

cs
Transport Layer 3-93
 TCP Congestion Control: details
sender limits transmission: How does sender
LastByteSent-LastByteAcked perceive congestion?
CongWin loss event = timeout or
Roughly, 3 duplicate acks
CongWin TCP sender reduces
rate = Bytes/sec
RTT rate (CongWin) after
CongWin is dynamic, function
loss event
of perceived network three mechanisms:
congestion AIMD
slow start

al

ep
conservative after
timeout events

itn
cs
Transport Layer 3-94
 TCP Slow Start
When connection begins, When connection begins,
CongWin = 1 MSS increase rate
Example: MSS = 500 exponentially fast until
bytes & RTT = 200 msec first loss event
initial rate = 20 kbps
available bandwidth may
be >> MSS/RTT
desirable to quickly ramp
up to respectable rate

al
ep
itn
cs
Transport Layer 3-95
TCP Slow Start (more)
When connection Host A Host B
begins, increase rate
exponentially until

RTT
first loss event:
double CongWin every
RTT
done by incrementing
CongWin for every ACK
received
Summary: initial rate
is slow but ramps up

al
ep
exponentially fast time

itn
cs
Transport Layer 3-96
 Refinement: inferring loss
After 3 dup ACKs:
CongWin is cut in half Philosophy:
window then grows
linearly 3 dup ACKs indicates
But after timeout event: network capable of
delivering some segments
CongWin instead set to
timeout indicates a
1 MSS;
more alarming
window then grows congestion scenario
exponentially

al
to a threshold, then

ep
grows linearly

itn
cs
Transport Layer 3-97
Refinement
Q: When should the
exponential
increase switch to
linear?
A: When CongWin
gets to 1/2 of its
value before
timeout.

Implementation:
Variable Threshold
At loss event, Threshold is
set to 1/2 of CongWin just

al
before loss event

ep
itn
cs
Transport Layer 3-98
Summary: TCP Congestion Control

When CongWin is below Threshold, sender in

slow-start phase, window grows exponentially.
When CongWin is above Threshold, sender is in
congestion-avoidance phase, window grows linearly.
When a triple duplicate ACK occurs, Threshold
set to CongWin/2 and CongWin set to
Threshold.

When timeout occurs, Threshold set to

al
CongWin/2 and CongWin is set to 1 MSS.

ep
itn
cs
Transport Layer 3-99
TCP sender congestion control
State Event TCP Sender Action Commentary
Slow Start ACK receipt CongWin = CongWin + MSS, Resulting in a doubling of
(SS) for previously If (CongWin > Threshold) CongWin every RTT
unacked set state to Congestion
data Avoidance
Congestion ACK receipt CongWin = CongWin+MSS * Additive increase, resulting
Avoidance for previously (MSS/CongWin) in increase of CongWin by
(CA) unacked 1 MSS every RTT
data
SS or CA Loss event Threshold = CongWin/2, Fast recovery,
detected by CongWin = Threshold, implementing multiplicative
triple Set state to Congestion decrease. CongWin will not
duplicate Avoidance drop below 1 MSS.
ACK
SS or CA Timeout Threshold = CongWin/2, Enter slow start
CongWin = 1 MSS,
Set state to Slow Start

al
SS or CA Duplicate Increment duplicate ACK count CongWin and Threshold not

ep
ACK for segment being acked changed

itn
cs
Transport Layer 3-100
TCP throughput
Whats the average throughout of TCP as a
function of window size and RTT?
Ignore slow start
Let W be the window size when loss occurs.
When window is W, throughput is W/RTT
Just after loss, window drops to W/2,
throughput to W/2RTT.
Average throughout: .75 W/RTT

al
ep
itn
cs
Transport Layer 3-101
TCP Futures: TCP over long, fat pipes

Example: 1500 byte segments, 100ms RTT, want 10

Gbps throughput
Requires window size W = 83,333 in-flight
segments
Throughput in terms of loss rate:

1.22 MSS
RTT L
L = 210-10 Wow

al
New versions of TCP for high-speed

ep
itn
cs
Transport Layer 3-102
TCP Fairness
Fairness goal: if K TCP sessions share same
bottleneck link of bandwidth R, each should have
average rate of R/K

TCP connection 1

bottleneck
TCP
router
connection 2
capacity R

al
ep
itn
cs
Transport Layer 3-103
Why is TCP fair?
Two competing sessions:
Additive increase gives slope of 1, as throughout increases
multiplicative decrease decreases throughput proportionally

R equal bandwidth share

loss: decrease window by factor of 2

congestion avoidance: additive increase
loss: decrease window by factor of 2
congestion avoidance: additive increase

al
ep
Connection 1 throughput R

itn
cs
Transport Layer 3-104
Fairness (more)
Fairness and UDP
Multimedia apps often
do not use TCP
do not want rate
throttled by congestion
control
Instead use UDP:
pump audio/video at
constant rate, tolerate
packet loss
Research area: TCP
friendly
Fairness and parallel TCP
connections
nothing prevents app from
opening parallel
connections between 2
hosts.
Web browsers do this
Example: link of rate R
supporting 9 connections;
new app asks for 1 TCP, gets
rate R/10
new app asks for 11 TCPs,
al
gets R/2 !
ep
itn
cs
Transport Layer 3-105
Chapter 3: Summary
principles behind transport
layer services:
multiplexing,
demultiplexing
reliable data transfer
flow control Next:
congestion control leaving the network
instantiation and edge (application,
implementation in the transport layers)
Internet into the network

al
UDP core

ep
TCP

itn
cs
Transport Layer 3-106