Chapter 3

Transport Layer

A note on the use of these ppt slides: Computer

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, Networking: A
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 Top Down
ask the following:
 If you use these slides (e.g., in a class) that you mention their source Approach
(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
6th edition
from (or perhaps identical to) our slides, and note our copyright of this Jim Kurose, Keith Ross
material. Addison-Wesley
Thanks and enjoy! JFK/KWR March 2012
All material copyright 1996-2013
J.F Kurose and K.W. Ross, All Rights Reserved

Transport Layer 3-1

Chapter 3: Transport Layer
our goals:
 understand  learn about Internet
principles behind transport layer
transport layer protocols:
services:  UDP: connectionless
 multiplexing, transport
demultiplexing  TCP: connection-
 reliable data oriented reliable
transfer transport
 flow control  TCP congestion control
 congestion control

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
3.3 connectionless  flow control
transport: UDP  connection
management
3.4 principles of
reliable data 3.6 principles of
transfer congestion control
3.7 TCP congestion
control

Transport Layer 3-3

Transport services and protocols
application
transport
 provide logical network
communication between app data link
physical
processes running on
different hosts

lo
gi
 transport protocols run in

ca
end systems

le
nd
 send side: breaks app

-e
d n
messages into segments,

tra
passes to network layer

ns
po
 rcv side: reassembles

t r
segments into messages, application
passes to app layer transport
network
 more than one transport data link
physical
protocol available to apps
 Internet: TCP and UDP

Transport Layer 3-4

Transport vs. network layer
 network layer: household analogy:
logical
communication 12 kids in Ann’ s house
sending letters to 12 kids
between hosts in Bill’ s house:
 transport layer:  hosts = houses

logical  processes = kids

communication  app messages = letters

in envelopes
between  transport protocol = Ann
processes and Bill who demux to in-
 relies on, house siblings
enhances,  network-layer protocol =
network layer postal service
services
Transport Layer 3-5
Internet transport-layer protocols
application
 reliable, in-order transport
network

delivery (TCP) data link

physical
network
 congestion control network data link

lo
data link physical

gi
physical
 flow control

ca
network

le
data link

nd
 connection setup physical

-e
n
network

d
unreliable, unordered

tra
data link
 physical

ns
delivery: UDP

po
network
data link

r t
physical
 no-frills extension of network
data link application

“ best-effort” IP physical
network
data link
transport
network
data link
services not
physical
 physical

available:
 delay guarantees
 bandwidth guarantees
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
3.3 connectionless  flow control
transport: UDP  connection
management
3.4 principles of
reliable data 3.6 principles of
transfer congestion control
3.7 TCP congestion
control

Transport Layer 3-7

Multiplexing/demultiplexing
multiplexing at sender:
handle data from multiple demultiplexing at receiver:
sockets, add transport use header info to deliver
header (later used for received segments to correct
demultiplexing) socket

application

application P1 P2 application socket

P3 transport P4
process
transport network transport
network link network
link physical link
physical physical

Transport Layer 3-8

How demultiplexing works
 host receives IP datagrams 32 bits
 each datagram has source IP
address, destination IP address source port # dest port #
 each datagram carries one
transport-layer segment
 each segment has source,
destination port number other header fields
 host uses IP addresses & port
numbers to direct segment to
appropriate socket
application
data
(payload)

TCP/UDP segment format

Transport Layer 3-9

Connectionless demultiplexing
 recall: created socket has  recall: when creating
host-local port #: datagram to send into
DatagramSocket mySocket1
= new
UDP socket, must
DatagramSocket(12534); specify
 destination IP address
 destination port #
 when host receives IP datagrams with
UDP segment: same dest. port #, but
 checks destination port different source IP
# in segment addresses and/or
source port numbers
 directs UDP segment to will be directed to same
socket with that port # socket at dest

Transport Layer 3-10

Connectionless demux:
example DatagramSocket serverSocket
= new DatagramSocket
(6428);
DatagramSocket DatagramSocket
mySocket2 = new mySocket1 = new
DatagramSocket DatagramSocket
(9157); application
(5775);
application P1 application
P3 P4
transport
transport transport
network
network link network
link physical link
physical physical

source port: 6428 source port: ?

dest port: 9157 dest port: ?

source port: 9157 source port: ?

dest port: 6428 dest port: ?
Transport Layer 3-11
Connection-oriented demux
 TCP socket identified  server host may
by 4-tuple: support many
 source IP address simultaneous TCP
 source port number sockets:
 dest IP address  each socket identified
 dest port number by its own 4-tuple
 demux: receiver uses
 web servers have
all four values to different sockets for
direct segment to each connecting client
appropriate socket  non-persistent HTTP
will have different
socket for each request

Transport Layer 3-12

Connection-oriented demux:
example
application
application P4 P5 P6 application
P3 P2 P3
transport
transport transport
network
network link network
link physical link
physical server: physical
IP
address
B
host: IP source IP,port: B,80 host: IP
address dest IP,port: A,9157 source IP,port: C,5775 address
A dest IP,port: B,80 C
source IP,port: A,9157
dest IP, port: B,80
source IP,port: C,9157
dest IP,port: B,80
three segments, all destined to IP address: B,
dest port: 80 are demultiplexed to different sockets Transport Layer 3-13
Connection-oriented demux:
example
threaded server
application
application application
P4
P3 P2 P3
transport
transport transport
network
network link network
link physical link
physical server: physical
IP
address
B
host: IP source IP,port: B,80 host: IP
address dest IP,port: A,9157 source IP,port: C,5775 address
A dest IP,port: B,80 C
source IP,port: A,9157
dest IP, port: B,80
source IP,port: C,9157
dest IP,port: B,80

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
3.3 connectionless  flow control
transport: UDP  connection
management
3.4 principles of
reliable data 3.6 principles of
transfer congestion control
3.7 TCP congestion
control

Transport Layer 3-15

UDP: User Datagram Protocol [RFC
768]
 “ no frills,” “ bare bones”  UDP use:
Internet transport protocol
 streaming multimedia
 “ best effort” service, UDP
segments may be: apps (loss tolerant,
 lost rate sensitive)
 delivered out-of-order  DNS
to app  SNMP
 connectionless:
 no handshaking
 reliable transfer over
between UDP sender, UDP:
receiver  add reliability at
 each UDP segment application layer
handled independently
of others  application-specific
error recovery!

Transport Layer 3-16

UDP: segment header
length, in bytes of
32 bits UDP segment,
source port # dest port # including header

length checksum
why is there a UDP?
 no connection
application establishment (which
data can add delay)
(payload)  simple: no connection
state at sender, receiver
 small header size
 no congestion control:
UDP segment format UDP can blast away as
fast as desired

Transport Layer 3-17

UDP checksum
Goal: detect “ errors” (e.g., flipped bits) in
transmitted segment
sender: receiver:
 treat segment contents,  compute checksum of
including header fields, received segment
as sequence of 16-bit  check if computed
integers checksum equals checksum
 checksum: addition (one field value:
’ s complement sum) of
segment contents  NO - error detected
 sender puts checksum  YES - no error detected.
value into UDP But maybe errors
checksum field nonetheless? More later
….

Transport Layer 3-18

Internet checksum: example
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

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

Note: when adding numbers, a carryout from the most

significant bit needs to be added to the result

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
3.3 connectionless  flow control
transport: UDP  connection
management
3.4 principles of
reliable data 3.6 principles of
transfer congestion control
3.7 TCP congestion
control

Transport Layer 3-20

Principles of reliable data
transfer
 important in application, transport, link layers
 top-10 list of important networking topics!

 characteristics of unreliable channel will determine complexity of reliable

data transfer protocol (rdt)

Transport Layer 3-21

Principles of reliable data
transfer
 important in application, transport, link layers
 top-10 list of important networking topics!

 characteristics of unreliable channel will determine complexity of reliable

data transfer protocol (rdt)

Transport Layer 3-22

Principles of reliable data
transfer
 important in application, transport, link layers
 top-10 list of important networking topics!

 characteristics of unreliable channel will determine complexity of reliable

data transfer protocol (rdt)

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

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

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

Transport Layer 3-24

 Reliable data transfer: getting started
we’ ll:
 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 uniquely state state
determined by next 1 event
event 2
actions

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 reads 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)

sender receiver

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
 How
senderdo retransmits
humans pktrecover
on receiptfrom
of NAK“ errors”
 new mechanisms in rdt2.0 (beyond rdt1.0):
during conversation?
 error detection
 receiver feedback: control msgs (ACK,NAK) rcvr-
>sender

Transport Layer 3-27

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
 feedback: control msgs (ACK,NAK) from receiver to
sender

Transport Layer 3-28

rdt2.0: FSM specification
rdt_send(data)
sndpkt = 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

call from
sender below

rdt_rcv(rcvpkt) &&
notcorrupt(rcvpkt)
extract(rcvpkt,data)
deliver_data(data)
udt_send(ACK)

Transport Layer 3-29

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
 call from
below

rdt_rcv(rcvpkt) &&
notcorrupt(rcvpkt)
extract(rcvpkt,data)
deliver_data(data)
udt_send(ACK)

Transport Layer 3-30

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
 call from
below

rdt_rcv(rcvpkt) &&
notcorrupt(rcvpkt)
extract(rcvpkt,data)
deliver_data(data)
udt_send(ACK)

Transport Layer 3-31

rdt2.0 has a fatal flaw!
what happens if handling duplicates:
ACK/NAK  sender retransmits current
corrupted? pkt if ACK/NAK corrupted
 sender doesn’ t know
 sender adds sequence
number to each pkt
what happened at
receiver!
 receiver discards (doesn’ t
deliver up) duplicate pkt
 can’ t just retransmit:
possible duplicate

stop and wait

sender sends one
packet,
then waits for receiver
response
Transport Layer 3-32
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)


Wait for Wait for
ACK or call 1 from
rdt_rcv(rcvpkt) && NAK 1 above
( corrupt(rcvpkt) ||
isNAK(rcvpkt) ) rdt_send(data)

udt_send(sndpkt) sndpkt = make_pkt(1, data, checksum)

udt_send(sndpkt)

Transport Layer 3-33

 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)

extract(rcvpkt,data)
deliver_data(data)
sndpkt = make_pkt(ACK, chksum)
udt_send(sndpkt)

Transport Layer 3-34

rdt2.1: discussion
sender: receiver:
 seq # added to pkt
 must check if
 two seq. #’ s (0,1) will
received packet is
suffice. Why? duplicate
 must check if received
ACK/NAK corrupted  state indicates
whether 0 or 1 is
 twice as many states
expected pkt seq #
 state must “ remember”
whether “ expected” pkt  note: receiver can
should have seq # of 0 not know if its last
or 1
ACK/NAK received
OK at sender

Transport Layer 3-35

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

Transport Layer 3-36

 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) || 
has_seq1(rcvpkt)) Wait for receiver FSM
0 from
udt_send(sndpkt) below fragment
rdt_rcv(rcvpkt) && notcorrupt(rcvpkt)
&& has_seq1(rcvpkt)
extract(rcvpkt,data)
deliver_data(data)
sndpkt = make_pkt(ACK1, chksum)
udt_send(sndpkt) Transport Layer 3-37
rdt3.0: channels with errors and loss

new assumption: approach: sender waits

underlying channel “ reasonable” amount
can also lose of time for ACK
packets (data,
 retransmits if no ACK
received in this time
ACKs)  if pkt (or ACK) just
 checksum, seq. #, delayed (not lost):
ACKs,  retransmission will be
retransmissions will duplicate, but seq. #’ s
already handles this
be of help … but not  receiver must specify
enough seq # of pkt being
ACKed
 requires countdown timer

Transport Layer 3-38

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 
 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) 
rdt_rcv(rcvpkt) &&
sndpkt = make_pkt(1, data, checksum)
( corrupt(rcvpkt) || udt_send(sndpkt)
isACK(rcvpkt,0) ) start_timer


Transport Layer 3-39

 rdt3.0 in action
sender receiver sender receiver
send pkt0 pkt0 send pkt0 pkt0
rcv pkt0 rcv pkt0
ack0 send ack0 ack0 send ack0
rcv ack0 rcv ack0
send pkt1 pkt1 send pkt1 pkt1
rcv pkt1 X
ack1 send ack1 loss
rcv ack1
send pkt0 pkt0
rcv pkt0 timeout
ack0 send ack0 resend pkt1 pkt1
rcv pkt1
ack1 send ack1
rcv ack1
send pkt0 pkt0
(a) no loss rcv pkt0
ack0 send ack0

(b) packet loss

Transport Layer 3-40
 rdt3.0 in action
sender receiver
sender receiver send pkt0 pkt0
send pkt0 pkt0 rcv pkt0
ack0 send ack0
rcv pkt0
send ack0 rcv ack0
ack0 send pkt1 pkt1
rcv ack0 rcv pkt1
send pkt1 pkt1
rcv pkt1 send ack1
ack1 ack1
send ack1
X
loss timeout
resend pkt1 pkt1
rcv pkt1
timeout
resend pkt1 pkt1 rcv ack1 pkt0 (detect duplicate)
rcv pkt1 send pkt0 send ack1
(detect duplicate) ack1
ack1 send ack1 rcv ack1 rcv pkt0
rcv ack1 ack0 send ack0
pkt0 send pkt0 pkt0
send pkt0 rcv pkt0
rcv pkt0 ack0 (detect duplicate)
ack0 send ack0 send ack0

(c) ACK loss (d) premature timeout/ delayed ACK

Transport Layer 3-41

Performance of rdt3.0
 rdt3.0 is correct, but performance stinks
 e.g.: 1 Gbps link, 15 ms prop. delay, 8000 bit packet:

L 8000 bits
Dtrans = R = = 8 microsecs
109 bits/sec

 U sender: utilization – fraction of time sender busy

sending L/R .008
U = 0.00027
sender = =
30.008
RTT + L / R
 if RTT=30 msec, 1KB pkt every 30 msec: 33kB/sec
thruput over 1 Gbps link
 network protocol limits use of physical resources!
Transport Layer 3-42
 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

U L/R .008
sender = = = 0.00027
RTT + L / R 30.008

Transport Layer 3-43

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

 two generic forms of pipelined protocols: go-

Back-N, selective repeat
Transport Layer 3-44
 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
3-packet pipelining increases
utilization by a factor of 3!

U 3L / R .0024
sender = = = 0.00081
RTT + L / R 30.008

Transport Layer 3-45

Pipelined protocols: overview
Go-back-N: Selective Repeat:
 sender can have up  sender can have up to
to N unacked packets N unack’ ed packets in
in pipeline pipeline
 receiver only sends  rcvr sends individual
cumulative ack ack for each packet
 doesn’ t ack packet if
there’ s a gap
 sender has timer for  sender maintains timer
oldest unacked for each unacked
packet packet
 when timer expires,  when timer expires,
retransmit all unacked retransmit only that
packets unacked packet

Transport Layer 3-46

Go-Back-N: sender
 k-bit seq # in pkt header
 “ window” of up to N, consecutive unack’ ed pkts allowed

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

“ cumulative ACK”
 may receive duplicate ACKs (see receiver)
 timer for oldest in-flight pkt
 timeout(n): retransmit packet n and all higher seq # pkts
in window
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++
}
 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) &&
notcorrupt(rcvpkt)
base = getacknum(rcvpkt)+1
If (base == nextseqnum)
stop_timer
else
start_timer
Transport Layer 3-48
GBN: receiver extended FSM
default
udt_send(sndpkt) rdt_rcv(rcvpkt)
&& notcurrupt(rcvpkt)
 && 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:
 discard (don’ t buffer): no receiver buffering!
 re-ACK pkt with highest in-order seq #
Transport Layer 3-49
 GBN in action
sender window (N=4) sender receiver
012345678 send pkt0
012345678 send pkt1
send pkt2 receive pkt0, send ack0
012345678
send pkt3 Xloss receive pkt1, send ack1
012345678
(wait)
receive pkt3, discard,
012345678 rcv ack0, send pkt4 (re)send ack1
012345678 rcv ack1, send pkt5 receive pkt4, discard,
(re)send ack1
ignore duplicate ACK receive pkt5, discard,
(re)send ack1
pkt 2 timeout
012345678 send pkt2
012345678 send pkt3
012345678 send pkt4 rcv pkt2, deliver, send ack2
012345678 send pkt5 rcv pkt3, deliver, send ack3
rcv pkt4, deliver, send ack4
rcv pkt5, deliver, send ack5

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
 limits seq #s of sent, unACKed pkts

Transport Layer 3-51

Selective repeat: sender, receiver
windows

Transport Layer 3-52

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

Transport Layer 3-53

 Selective repeat in action
sender window (N=4) sender receiver
012345678 send pkt0
012345678 send pkt1
send pkt2 receive pkt0, send ack0
012345678
send pkt3 Xloss receive pkt1, send ack1
012345678
(wait)
receive pkt3, buffer,
012345678 rcv ack0, send pkt4 send ack3
012345678 rcv ack1, send pkt5 receive pkt4, buffer,
send ack4
record ack3 arrived receive pkt5, buffer,
send ack5
pkt 2 timeout
012345678 send pkt2
012345678 record ack4 arrived
012345678 rcv pkt2; deliver pkt2,
record ack5 arrived
012345678 pkt3, pkt4, pkt5; send ack2

Q: what happens when ack2 arrives?

Transport Layer 3-54

 sender window receiver window
Selective repeat: (after receipt) (after receipt)

dilemma 0123012 pkt0

pkt1 0123012
0123012
0123012 pkt2 0123012
example: 0123012
0123012 pkt3
 seq #’ s: 0, 1, 2, 3 0123012
X
 window size=3 pkt0 will accept packet
with seq number 0
(a) no problem
 receiver sees no
difference in two receiver can’t see sender side.
scenarios! receiver behavior identical in both cases!
something’s (very) wrong!
 duplicate data
accepted as new in 0123012 pkt0
(b) 0123012 pkt1 0123012
0123012 pkt2 0123012
X 0123012
Q: what relationship X
timeout
between seq # size retransmit pkt0 X
and window size to 0123012 pkt0
will accept packet
avoid problem in (b) oops!
with seq number 0
(b)?
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
3.3 connectionless  flow control
transport: UDP  connection
management
3.4 principles of
reliable data 3.6 principles of
transfer congestion control
3.7 TCP congestion
control

Transport Layer 3-56

TCP: Overview RFCs: 793,1122,1323, 2018,
2581

 point-to-point:  full duplex data:

 one sender, one  bi-directional data flow
in same connection
receiver
 MSS: maximum
 reliable, in-order byte segment size
steam:  connection-oriented:
 no “ message  handshaking (exchange
boundaries” of control msgs) inits
sender, receiver state
 pipelined: before data exchange
 TCP congestion and  flow controlled:
flow control set  sender will not
window size overwhelm receiver

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
UAP R S F receive window
(generally not used) # bytes
checksum Urg data pointer
rcvr willing
RST, SYN, FIN: to accept
options (variable length)
connection estab
(setup, teardown
commands)
application
Internet data
checksum (variable length)
(as in UDP)

Transport Layer 3-58

TCP seq. numbers, ACKs
outgoing segment from sender
sequence numbers: source port # dest port #
sequence number
 byte stream “ number” acknowledgement number
of first byte in rwnd
segment’ s data checksum urg pointer

acknowledgements: window size

N
 seq # of next byte
expected from other
side sender sequence number space
 cumulative ACK
sent sent, not- usable not
Q: how receiver handles ACKed yet ACKed but not usable
out-of-order segments (“in-flight”) yet sent
 A: TCP spec doesn’ t incoming segment to sender
say, - up to source port # dest port #
implementor sequence number
acknowledgement number
A rwnd
checksum urg pointer

Transport Layer 3-59

TCP seq. numbers, ACKs
Host A Host B

User
types
‘C’
Seq=42, ACK=79, data = ‘C’
host ACKs
receipt of
‘C’, echoes
Seq=79, ACK=43, data = ‘C’ back ‘C’
host ACKs
receipt
of echoed
‘C’ Seq=43, ACK=80

simple telnet scenario

Transport Layer 3-60

TCP round trip time, timeout
Q: how to set TCP Q: how to estimate
timeout value? RTT?
 longer than RTT  SampleRTT: measured
 but RTT varies time from segment
transmission until ACK
 too short: receipt
premature timeout,  ignore retransmissions
unnecessary
retransmissions  SampleRTT will vary,
want estimated RTT
 too long: slow “ smoother”
reaction to segment  average several recent
loss measurements, not just
current SampleRTT

Transport Layer 3-61

TCP round trip time, timeout
EstimatedRTT = (1- )*EstimatedRTT + *SampleRTT
 exponential weighted moving average
 influence of past sample decreases
exponentially fast RTT: gaia.cs.umass.edu to fantasia.eurecom.fr

 typical value:  = 0.125 350

RTT: gaia.cs.umass.edu to fantasia.eurecom.fr

300
(milliseconds)
RTT

250
RTT (milliseconds)

200

sampleRTT
150

EstimatedRTT

100
1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106
time (seconnds)
time (seconds) Transport Layer 3-62
SampleRTT Estimated RTT
TCP round trip time, timeout
 timeout interval: EstimatedRTT plus “ safety
margin”
 large variation in EstimatedRTT -> larger safety margin
 estimate SampleRTT deviation from EstimatedRTT:
DevRTT = (1-)*DevRTT +
*|SampleRTT-EstimatedRTT|
(typically,  = 0.25)

TimeoutInterval = EstimatedRTT + 4*DevRTT

estimated RTT “safety margin”

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
3.3 connectionless  flow control
transport: UDP  connection
management
3.4 principles of
reliable data 3.6 principles of
transfer congestion control
3.7 TCP congestion
control

Transport Layer 3-64

TCP reliable data transfer
 TCP creates rdt
service on top of IP’ s
unreliable service
 pipelined segments
 cumulative acks let’ s initially consider
 single retransmission
simplified TCP
timer
sender:
 ignore duplicate acks
 retransmissions  ignore flow control,
triggered by: congestion control
 timeout events
 duplicate acks

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 ack acknowledges
 start timer if not
previously unacked
already running segments
 think of timer as for  update what is known
oldest unacked to be ACKed
segment  start timer if there are
 expiration interval: still unacked
TimeOutInterval segments

Transport Layer 3-66

TCP sender (simplified)
data received from application above
create segment, seq. #: NextSeqNum
pass segment to IP (i.e., “send”)
NextSeqNum = NextSeqNum + length(data)
if (timer currently not running)
 start timer
NextSeqNum = InitialSeqNum wait
SendBase = InitialSeqNum for
event timeout
retransmit not-yet-acked segment
with smallest seq. #
start timer
ACK received, with ACK field value y
if (y > SendBase) {
SendBase = y
/* SendBase–1: last cumulatively ACKed byte */
if (there are currently not-yet-acked segments)
start timer
else stop timer
} Transport Layer 3-67
TCP: retransmission scenarios
Host A Host B Host A Host B

SendBase=92
Seq=92, 8 bytes of data Seq=92, 8 bytes of data

Seq=100, 20 bytes of data

timeo

timeo
ACK=100
ut

ut
X
ACK=100
ACK=120

Seq=92, 8 bytes of data Seq=92, 8

SendBase=100 bytes of data
SendBase=120
ACK=100
ACK=120

SendBase=120

lost ACK scenario premature timeout

Transport Layer 3-68
TCP: retransmission scenarios
Host A Host B

Seq=92, 8 bytes of data

Seq=100, 20 bytes of data

ACK=100
timeo

X
ut

ACK=120

Seq=120, 15 bytes of data

cumulative ACK
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

arrival of segment that immediate send ACK, provided that

partially or completely fills gap segment starts at lower end of gap

Transport Layer 3-70

TCP fast retransmit
 time-out period
often relatively long: TCP fast retransmit
 long delay before if sender receives 3
resending lost packet
ACKs for same data
 detect lost segments
via duplicate ACKs. (“ triple
(“ triple duplicate
duplicate ACKs
 sender often sends ”ACKs”
), ), resend
many segments unacked segment
back-to-back with smallest seq #
 if segment is lost,  likely that unacked
there will likely be segment lost, so don
many duplicate ’ t wait for timeout
ACKs.

Transport Layer 3-71

TCP fast retransmit
Host A Host B

Seq=92, 8 bytes of data

Seq=100, 20 bytes of data
X

ACK=100
timeo

ACK=100
ut

ACK=100
ACK=100
Seq=100, 20 bytes of data

fast retransmit after sender

receipt of triple duplicate ACK
Transport Layer 3-72
Chapter 3 outline
3.1 transport-layer 3.5 connection-oriented
services transport: TCP
3.2 multiplexing and  segment structure
demultiplexing  reliable data transfer
3.3 connectionless  flow control
transport: UDP  connection
management
3.4 principles of
reliable data 3.6 principles of
transfer congestion control
3.7 TCP congestion
control

Transport Layer 3-73

TCP flow control
application
application may process
remove data from application
TCP socket buffers ….
TCP socket OS
receiver buffers
… slower than TCP
receiver is delivering
(sender is sending) TCP
code

IP
flow control code
receiver controls sender, so
sender won’ t overflow
receiver’ s buffer by from sender
transmitting too much, too fast
receiver protocol stack

Transport Layer 3-74

TCP flow control
 receiver “ advertises” free
buffer space by including to application process
rwnd value in TCP
header of receiver-to-
sender segments RcvBuffer buffered data
 RcvBuffer size set via
socket options (typical rwnd free buffer space
default is 4096 bytes)
 many operating systems
autoadjust RcvBuffer TCP segment payloads
 sender limits amount of
unacked (“ in-flight” ) data receiver-side buffering
to receiver’ s rwnd value
 guarantees receive buffer
will not overflow
Transport Layer 3-75
Chapter 3 outline
3.1 transport-layer 3.5 connection-oriented
services transport: TCP
3.2 multiplexing and  segment structure
demultiplexing  reliable data transfer
3.3 connectionless  flow control
transport: UDP  connection
management
3.4 principles of
reliable data 3.6 principles of
transfer congestion control
3.7 TCP congestion
control

Transport Layer 3-76

Connection Management
before exchanging data, sender/receiver
“ handshake” :
 agree to establish connection (each knowing the other
willing to establish connection)
 agree on connection parameters
application application

connection state: ESTAB connection state: ESTAB

connection variables: connection Variables:
seq # client-to-server seq # client-to-server
server-to-client server-to-client
rcvBuffer size rcvBuffer size
at server,client at server,client

network network

Socket clientSocket = Socket connectionSocket =

newSocket("hostname","port welcomeSocket.accept();
number");
Transport Layer 3-77
Agreeing to establish a connection

2-way handshake:
Q: will 2-way handshake
always work in network?
 variable delays
Let’s talk  retransmitted messages (e.g.
ESTAB req_conn(x)) due to message
OK loss
ESTAB  message reordering
 can’ t “ see” other side

choose x
req_conn(x)
ESTAB
acc_conn(x)
ESTAB

Transport Layer 3-78

Agreeing to establish a connection
2-way handshake failure scenarios:

choose x choose x
req_conn(x) req_conn(x)
ESTAB ESTAB
retransmit acc_conn(x) retransmit acc_conn(x)
req_conn(x) req_conn(x)

ESTAB ESTAB
data(x+1) accept
req_conn(x)
retransmit data(x+1)
data(x+1)
connection connection
client x completes server x completes server
client
terminates forgets x terminates forgets x
req_conn(x)

ESTAB ESTAB
data(x+1) accept
half open connection! data(x+1)
(no client!)
Transport Layer 3-79
 TCP 3-way handshake

client state server state

LISTEN LISTEN
choose init seq num, x
send TCP SYN msg
SYNSENT SYNbit=1, Seq=x
choose init seq num, y
send TCP SYNACK
msg, acking SYN SYN RCVD
SYNbit=1, Seq=y
ACKbit=1; ACKnum=x+1
received SYNACK(x)
ESTAB indicates server is live;
send ACK for SYNACK;
this segment may contain ACKbit=1, ACKnum=y+1
client-to-server data
received ACK(y)
indicates client is live
ESTAB

Transport Layer 3-80

TCP 3-way handshake:
FSM
closed

Socket connectionSocket =
welcomeSocket.accept();

 Socket clientSocket =
SYN(x) newSocket("hostname","port
number");
SYNACK(seq=y,ACKnum=x+1)
create new socket for listen SYN(seq=x)
communication back to client

SYN SYN
rcvd sent

SYNACK(seq=y,ACKnum=x+1)
ESTAB ACK(ACKnum=y+1)
ACK(ACKnum=y+1)


Transport Layer 3-81

TCP: closing a connection
 client, server each close their side of
connection
 send TCP segment with FIN bit = 1
 respond to received FIN with ACK
 on receiving FIN, ACK can be combined with own
FIN
 simultaneous FIN exchanges can be handled

Transport Layer 3-82

TCP: closing a connection
client state server state
ESTAB ESTAB
clientSocket.close()
FIN_WAIT_1 can no longer FINbit=1, seq=x
send but can
receive data CLOSE_WAIT
ACKbit=1; ACKnum=x+1
can still
FIN_WAIT_2 wait for server send data
close

LAST_ACK
FINbit=1, seq=y
TIMED_WAIT can no longer
send data
ACKbit=1; ACKnum=y+1
timed wait
for 2*max CLOSED
segment lifetime

CLOSED

Transport Layer 3-83

Chapter 3 outline
3.1 transport-layer 3.5 connection-oriented
services transport: TCP
3.2 multiplexing and  segment structure
demultiplexing  reliable data transfer
3.3 connectionless  flow control
transport: UDP  connection
management
3.4 principles of
reliable data 3.6 principles of
transfer congestion control
3.7 TCP congestion
control

Transport Layer 3-84

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!

Transport Layer 3-85

Causes/costs of congestion: scenario
1
original data: in throughput:out
 two senders, two
receivers Host A

 one router, infinite unlimited shared

buffers output link buffers

 output link capacity: R

 no retransmission
Host B

R/2

delay
out

in R/2 in R/2

 maximum per-connection  large delays as arrival rate,
throughput: R/2 in, approaches capacity
Transport Layer 3-86
Causes/costs of congestion: scenario
2
 one router, finite buffers
 sender retransmission of timed-out packet
 application-layer input = application-layer output:in = out
 transport-layer input includes retransmissions :in in
‘

in : original data

'in: original data, plus out
retransmitted data

Host A

finite shared output

Host B
link buffers
Transport Layer 3-87
Causes/costs of congestion: scenario
2
R/2
idealization: perfect
knowledge

out
 sender sends only when
router buffers available
in R/2

in : original data

copy 'in: original data, plus out
retransmitted data

A free buffer space!

finite shared output

Host B
link buffers
Transport Layer 3-88
Causes/costs of congestion: scenario
2
Idealization: known
loss packets can be
lost, dropped at router
due to full buffers
 sender only resends if
packet known to be lost

in : original data

copy out
'in: original data, plus
retransmitted data

A no buffer space!

Host B
Transport Layer 3-89
Causes/costs of congestion: scenario
2
Idealization: known R/2
loss packets can be
lost, dropped at router when sending at R/2,
due to full buffers some packets are

out
retransmissions but
 sender only resends if asymptotic goodput
packet known to be lost is still R/2 (why?)

in R/2

in : original data

out
'in: original data, plus
retransmitted data

A free buffer space!

Host B
Transport Layer 3-90
Causes/costs of congestion: scenario
2
Realistic: duplicates R/2
 packets can be lost,
dropped at router due to when sending at R/2,
some packets are

out
full buffers retransmissions
including duplicated
 sender times out that are delivered!
prematurely, sending two in R/2
copies, both of which are
delivered in
timeout
copy out
'in

A free buffer space!

Host B
Transport Layer 3-91
Causes/costs of congestion: scenario
2
Realistic: duplicates R/2
 packets can be lost,
dropped at router due to when sending at R/2,
some packets are

out
full buffers retransmissions
including duplicated
 sender times out that are delivered!
prematurely, sending two in R/2
copies, both of which are
delivered
“ costs” of congestion:
 more work (retrans) for given “ goodput”
 unneeded retransmissions: link carries multiple copies
of pkt
 decreasing goodput

Transport Layer 3-92

Causes/costs of congestion: scenario
3
 four senders Q: what happens as in and in’
increase ?
 multihop paths
A: as red in’ increases, all
 timeout/retransmit arriving blue pkts at upper
queue are dropped, blue
Host A
in : original throughput
data

 out
0
Host B
'in: original data, plus
retransmitted data
finite shared output
link buffers

Host D
Host C

Transport Layer 3-93

Causes/costs of congestion: scenario
3
C/2
out

in’ C/2

another “ cost” of congestion:

 when packet dropped, any “ upstream
transmission capacity used for that packet
was wasted!

Transport Layer 3-94

Approaches towards congestion
control
two broad approaches towards congestion control:

end-end network-assisted
congestion congestion control:
control:  routers provide
 no explicit feedback feedback to end
from network systems
 congestion inferred  single bit indicating
from end-system
observed loss, congestion (SNA,
delay DECbit, TCP/IP
 approach taken by ECN, ATM)
TCP  explicit rate for
sender to send at
Transport Layer 3-95
Case study: ATM ABR congestion
control
ABR: available bit RM (resource
rate: management) cells:
 sent by sender, interspersed
 “ elastic service” with data cells
 if sender’ s path  bits in RM cell set by switches
“ underloaded” : (“ network-assisted” )
 sender should use  NI bit: no increase in rate
(mild congestion)
available bandwidth
 CI bit: congestion
 if sender’ s path indication
congested:  RM cells returned to sender
 sender throttled to by receiver, with bits intact
minimum
guaranteed rate

Transport Layer 3-96

Case study: ATM ABR congestion
control
RM cell data cell

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

 congested switch may lower ER value in cell
 senders’ send rate thus max supportable rate on path
 EFCI bit in data cells: set to 1 in congested switch
 if data cell preceding RM cell has EFCI set, receiver sets
CI bit in returned RM cell

Transport Layer 3-97

Chapter 3 outline
3.1 transport-layer 3.5 connection-oriented
services transport: TCP
3.2 multiplexing and  segment structure
demultiplexing  reliable data transfer
3.3 connectionless  flow control
transport: UDP  connection
management
3.4 principles of
reliable data 3.6 principles of
transfer congestion control
3.7 TCP congestion
control

Transport Layer 3-98

 TCP congestion control: additive
increase multiplicative decrease
 approach: sender increases transmission rate
(window size), probing for usable bandwidth, until
loss occurs
 additive increase: increase cwnd by 1 MSS
every RTT until loss detected
 multiplicative decrease: cut cwnd in half after
additively increase window size …
loss …. until loss occurs (then cut window in half)
congestion window size
cwnd: TCP sender

AIMD saw tooth

behavior: probing
for bandwidth

time
Transport Layer 3-99
TCP Congestion Control:
details
sender sequence number space
cwnd TCP sending rate:
 roughly: send cwnd
bytes, wait RTT for
last byte last byte ACKS, then send
ACKed sent, not-
yet ACKed
sent more bytes
(“in-flight”)
cwnd
 sender limits transmission: rate ~
~ bytes/sec
RTT
LastByteSent- < cwnd
LastByteAcked
 cwnd is dynamic, function of
perceived network congestion

Transport Layer 3-100

TCP Slow Start
Host A Host B
 when connection
begins, increase rate
exponentially until first on e s e g m
e nt

RTT
loss event:
 initially cwnd = 1 MSS two segm
ents
 double cwnd every RTT
 done by incrementing
cwnd for every ACK four segm
ents
received
 summary: initial rate is
slow but ramps up
exponentially fast time

Transport Layer 3-101

TCP: detecting, reacting to
loss
 loss indicated by timeout:
 cwnd set to 1 MSS;
 window then grows exponentially (as in slow start) to threshold, then
grows linearly
 loss indicated by 3 duplicate ACKs: TCP RENO
 dup ACKs indicate network capable of delivering some segments
 cwnd is cut in half window then grows linearly
 TCP Tahoe always sets cwnd to 1 (timeout or 3 duplicate acks)

Transport Layer 3-102

TCP: switching from slow start to
CA
Q: when should the
exponential
increase switch to
linear?
A: when cwnd gets
to 1/2 of its value
before timeout.

Implementation:
 variable ssthresh
 on loss event, ssthresh
is set to 1/2 of cwnd just
before loss event

Transport Layer 3-103

Summary: TCP Congestion
Control New
New ACK!
ACK!
duplicate ACK
dupACKcount++ new ACK
new ACK
.
cwnd = cwnd + MSS (MSS/cwnd)
dupACKcount = 0
cwnd = cwnd+MSS transmit new segment(s), as allowed
dupACKcount = 0
 transmit new segment(s), as allowed
cwnd = 1 MSS
ssthresh = 64 KB cwnd > ssthresh
dupACKcount = 0 slow  congestion
start timeout avoidance
ssthresh = cwnd/2
cwnd = 1 MSS duplicate ACK
timeout dupACKcount = 0 dupACKcount++
ssthresh = cwnd/2 retransmit missing segment
cwnd = 1 MSS
dupACKcount = 0
retransmit missing segment New
timeout
ACK!
ssthresh = cwnd/2
cwnd = 1 New ACK
dupACKcount = 0
retransmit missing segment cwnd = ssthresh dupACKcount == 3
dupACKcount == 3 dupACKcount = 0
ssthresh= cwnd/2 ssthresh= cwnd/2
cwnd = ssthresh + 3 cwnd = ssthresh + 3
retransmit missing segment retransmit missing segment
fast
recovery
duplicate ACK
cwnd = cwnd + MSS
transmit new segment(s), as allowed

Transport Layer 3-104

TCP throughput
 avg. TCP thruput as function of window size,
RTT?
 ignore slow start, assume always data to send
 W: window size (measured in bytes) where loss occurs
 avg. window size (# in-flight bytes) is ¾ W
 avg. thruput is 3/4W per RTT
3 W
avg TCP thruput = bytes/sec
4 RTT

W/2

Transport Layer 3-105

TCP Futures: TCP over “ long, fat
pipes”
 example: 1500 byte segments, 100ms RTT,
want 10 Gbps throughput
 requires W = 83,333 in-flight segments
 throughput in terms of segment loss
probability, L [Mathis 1997]:
1.22 . MSS
TCP throughput =
RTT L

➜ to achieve 10 Gbps throughput, need a loss rate

of L = 2·10-10 – a very small loss rate!
 new versions of TCP for high-speed

Transport Layer 3-106

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
router
capacity R
TCP connection 2

Transport Layer 3-107

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
Connection 2 throughput

loss: decrease window by factor of 2

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

Connection 1 throughput R
Transport Layer 3-108
Fairness (more)
Fairness and UDP Fairness, parallel TCP
 multimedia apps connections
 application can open
often do not use
TCP multiple parallel
 do not want rate connections between two
throttled by hosts
congestion control  web browsers do this
 instead use UDP:  e.g., link of rate R with 9
 send audio/video at existing connections:
constant rate,  new app asks for 1 TCP, gets
tolerate packet loss rate R/10
 new app asks for 11 TCPs, gets
R/2

Transport Layer 3-109

Chapter 3: summary
 principles behind
transport layer services:
 multiplexing,
demultiplexing
 reliable data transfer next:
 leaving the
 flow control
 congestion control network “ edge”
(application,
 instantiation, transport layers)
implementation in the
Internet  into the network
 UDP “ core”
 TCP

Transport Layer 3-110