Transport layer (computer networks)

Chapter 3
Transport Layer
A note on the use of these ppt 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.
Thanks and enjoy! JFK/KWR

Computer
Networking: A Top
Down Approach
6th edition
Jim Kurose, Keith Ross
Addison-Wesley
March 2012

All material copyright 1996-2012
J.F Kurose and K.W. Ross, All Rights Reserved
Transport Layer 3-1

Chapter 3: Transport Layer
our goals:


understand
principles behind
transport layer
services:
 multiplexing,
demultiplexing
 reliable data transfer
 flow control
 congestion control



learn about Internet
transport layer protocols:
 UDP: connectionless
transport
 TCP: connection-oriented
reliable transport
 TCP congestion control

Transport Layer 3-2

Chapter 3 outline
3.1 transport-layer
services
3.2 multiplexing and
demultiplexing
3.3 connectionless
transport: UDP
3.4 principles of reliable
data transfer

3.5 connection-oriented
transport: TCP
 segment structure
 flow control
 connection management

3.6 principles of congestion
control
3.7 TCP congestion control

Transport Layer 3-3

Transport services and protocols


rt
po
ns
tra



nd
-e
nd
le
ca
gi
lo



provide logical communication
between app processes
running on different hosts
transport protocols run in
end systems
 send side: breaks app
messages into segments,
passes to network layer
 rcv side: reassembles
segments into messages,
passes to app layer
more than one transport
protocol available to apps
 Internet: TCP and UDP

application
transport
network
data link
physical

application
transport
network
data link
physical

Transport Layer 3-4

Transport vs. network layer
network layer: logical
communication
between hosts
 transport layer:
logical
communication
between processes


 relies on, enhances,
network layer
services

household analogy:
12 kids in Ann’s house sending
letters to 12 kids in Bill’s
house:
 hosts = houses
 processes = kids
 app messages = letters in
envelopes
 transport protocol = Ann
and Bill who demux to inhouse siblings
 network-layer protocol =
postal service
Transport Layer 3-5

Internet transport-layer protocols


reliable, in-order
delivery (TCP)

unreliable, unordered
delivery: UDP
 no-frills extension of
“best-effort” IP



services not available:

network
data link
physical

network
data link
physical

network
data link
physical

network
data link
physical

rt
po
ns
tra



network
data link
physical

nd
-e
nd
le
ca
gi
lo

 flow control
 connection setup

application
transport
network
data link
physical

network
data link
physical

network
data link
physical

application
transport
network
data link
physical

 delay guarantees
 bandwidth guarantees
Transport Layer 3-6

Chapter 3 outline
3.1 transport-layer
services
demultiplexing
3.3 connectionless
transport: UDP
data transfer

transport: TCP
 flow control

control

Transport Layer 3-7

Multiplexing/demultiplexing
multiplexing at sender:
handle data from multiple
sockets, add transport header
(later used for demultiplexing)

demultiplexing at receiver:
use header info to deliver
received segments to correct
socket

application
application

P1

P2

application

P3

transport

P4

transport

network

transport

network

link

socket

network

link
physical

physical

process

link
physical

Transport Layer 3-8

How demultiplexing works


host receives IP datagrams
 each datagram has source IP
address, destination IP
address
 each datagram carries one
transport-layer segment
 each segment has source,
destination port number



host uses IP addresses &
port numbers to direct
segment to appropriate
socket

32 bits
source port #

dest port #

other header fields
application
data
(payload)
TCP/UDP segment format

Transport Layer 3-9

Connectionless demultiplexing


recall: created socket has
host-local port #:
DatagramSocket mySocket1
= new
DatagramSocket(12534);



when host receives UDP
segment:
 checks destination port #
in segment
 directs UDP segment to
socket with that port #



recall: when creating
datagram to send into
UDP socket, must specify
 destination IP address
 destination port #
IP datagrams with same
dest. port #, but different
source IP addresses
and/or source port
numbers will be directed
to same socket at dest
Transport Layer 3-10

Connectionless demux: example
DatagramSocket
mySocket2 = new
DatagramSocket
(9157);

DatagramSocket
serverSocket = new
DatagramSocket
(6428);
application

DatagramSocket
mySocket1 = new
DatagramSocket
(5775);

P1

application

application

P3

transport

transport

transport

network

network

network

link

link

link

physical

physical
source port: 6428
dest port: 9157

source port: 9157
dest port: 6428

P4

physical
source port: ?
dest port: ?
source port: ?
dest port: ?

Connection-oriented demux


TCP socket identified
by 4-tuple:
 source IP address
 source port number
 dest IP address
 dest port number



demux: receiver uses
all four values to direct
segment to appropriate
socket



server host may support
many simultaneous TCP
sockets:
 each socket identified by
its own 4-tuple



web servers have
different sockets for
each connecting client
 non-persistent HTTP will
have different socket for
each request


Connection-oriented demux: example
application
application

P4

P3

application

P6

transport

network

network

network

link

link

link

physical

physical

P3

P2

transport

transport

host: IP
address A

P5

server: IP
address B
source IP,port: B,80
dest IP,port: A,9157
source IP,port: A,9157
dest IP, port: B,80

three segments, all destined to IP address: B,
dest port: 80 are demultiplexed to different sockets

physical

source IP,port: C,5775
dest IP,port: B,80

host: IP
address C

dest IP,port: B,80

Connection-oriented demux: example
threaded server
application
application

P3

transport

network

network

network

link

link

link

physical

physical

server: IP
address B
source IP,port: B,80
dest IP,port: A,9157
source IP,port: A,9157
dest IP, port: B,80

P3

P2

transport

transport

host: IP
address A

application

P4

physical

dest IP,port: B,80

host: IP
address C

dest IP,port: B,80

Chapter 3 outline
3.1 transport-layer
services
demultiplexing
3.3 connectionless
transport: UDP
data transfer

transport: TCP
 flow control

control


UDP: User Datagram Protocol [RFC 768]






“no frills,” “bare bones”
Internet transport
protocol
“best effort” service, UDP
segments may be:
 lost
 delivered out-of-order
to app
connectionless:
 no handshaking
between UDP sender,
receiver
 each UDP segment
handled independently
of others



UDP use:
 streaming multimedia
apps (loss tolerant, rate
sensitive)
 DNS
 SNMP



reliable transfer over
UDP:
 add reliability at
application layer
 application-specific error
recovery!


UDP: segment header
32 bits
source port #

dest port #

length

checksum

length, in bytes of
UDP segment,
including header

why is there a UDP?


application
data
(payload)





UDP segment format



no connection
establishment (which can
add delay)
simple: no connection
state at sender, receiver
small header size
no congestion control:
UDP can blast away as
fast as desired

UDP checksum
Goal: detect “errors” (e.g., flipped bits) in transmitted
segment

sender:






treat segment contents,
including header fields,
as sequence of 16-bit
integers
checksum: addition
(one’s complement sum)
of segment contents
sender puts checksum
value into UDP
checksum field

receiver:




compute checksum of
received segment
check if computed
checksum equals checksum
field value:
 NO - error detected
 YES - no error detected.
But maybe errors
nonetheless? More later
….

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


Chapter 3 outline
3.1 transport-layer
services
demultiplexing
3.3 connectionless
transport: UDP
data transfer

transport: TCP
 flow control

control


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)

transfer






Reliable data transfer: getting started
rdt_send(): called from above,
(e.g., by app.). Passed data to
deliver to receiver upper layer

send
side

udt_send(): called by rdt,
to transfer packet over
unreliable channel to receiver

deliver_data(): called by
rdt to deliver data to upper

receive
side

rdt_rcv(): called when packet
arrives on rcv-side of channel

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
determined by next
event

state
1

event
actions

state
2


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
call from
above

rdt_send(data)
packet = make_pkt(data)
udt_send(packet)

sender

Wait for
call from
below

rdt_rcv(packet)
extract (packet,data)
deliver_data(data)

receiver

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
How do humans recover from “errors”
new mechanisms in rdt2.0 (beyond rdt1.0):
during conversation?
 error detection
 receiver feedback: control msgs (ACK,NAK) rcvr>sender


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

rdt2.0: FSM specification
rdt_send(data)
sndpkt = make_pkt(data, checksum)
udt_send(sndpkt)
rdt_rcv(rcvpkt) &&
isNAK(rcvpkt)
Wait for
Wait for
call from
ACK or
udt_send(sndpkt)
above
NAK
rdt_rcv(rcvpkt) && isACK(rcvpkt)
Λ

sender

receiver
rdt_rcv(rcvpkt) &&
corrupt(rcvpkt)
udt_send(NAK)
Wait for
call from
below
rdt_rcv(rcvpkt) &&
notcorrupt(rcvpkt)
extract(rcvpkt,data)
deliver_data(data)
udt_send(ACK)

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
call from
ACK or
udt_send(sndpkt)
above
NAK
Λ

rdt_rcv(rcvpkt) &&
corrupt(rcvpkt)
udt_send(NAK)
Wait for
call from
below
rdt_rcv(rcvpkt) &&
notcorrupt(rcvpkt)
deliver_data(data)
udt_send(ACK)

rdt2.0: error scenario
rdt_send(data)
snkpkt = make_pkt(data, checksum)
udt_send(sndpkt)
rdt_rcv(rcvpkt) &&
isNAK(rcvpkt)
Wait for
Wait for
call from
ACK or
udt_send(sndpkt)
above
NAK
Λ

rdt_rcv(rcvpkt) &&
corrupt(rcvpkt)
udt_send(NAK)
Wait for
call from
below
rdt_rcv(rcvpkt) &&
notcorrupt(rcvpkt)
deliver_data(data)
udt_send(ACK)

rdt2.0 has a fatal flaw!
what happens if
ACK/NAK corrupted?




sender doesn’t know
what happened at
receiver!
can’t just retransmit:
possible duplicate

handling duplicates:






sender retransmits
current pkt if ACK/NAK
corrupted
sender adds sequence
number to each pkt
receiver discards (doesn’t
deliver up) duplicate pkt

stop and wait
sender sends one packet,
then waits for receiver
response

rdt2.1: sender, handles garbled ACK/NAKs
rdt_send(data)
sndpkt = make_pkt(0, data, checksum)
udt_send(sndpkt)
rdt_rcv(rcvpkt) &&

rdt_rcv(rcvpkt)
&& notcorrupt(rcvpkt)
&& isACK(rcvpkt)

Wait for
call 0 from
above

rdt_rcv(rcvpkt)
&& isACK(rcvpkt)

Λ
rdt_rcv(rcvpkt) &&
( corrupt(rcvpkt) ||
isNAK(rcvpkt) )
udt_send(sndpkt)

isNAK(rcvpkt) )
udt_send(sndpkt)

Wait for
ACK or
NAK 0

Λ

Wait for
ACK or
NAK 1

Wait for
call 1 from
above

rdt_send(data)
udt_send(sndpkt)


rdt2.1: receiver, handles garbled ACK/NAKs
rdt_rcv(rcvpkt) && notcorrupt(rcvpkt)
&& has_seq0(rcvpkt)

rdt_rcv(rcvpkt) && (corrupt(rcvpkt)

deliver_data(data)
sndpkt = make_pkt(ACK, chksum)
udt_send(sndpkt)
rdt_rcv(rcvpkt) && (corrupt(rcvpkt)

sndpkt = make_pkt(NAK, chksum)
udt_send(sndpkt)
rdt_rcv(rcvpkt) &&
not corrupt(rcvpkt) &&
has_seq1(rcvpkt)
udt_send(sndpkt)

sndpkt = make_pkt(NAK, chksum)
udt_send(sndpkt)
Wait for
0 from
below

Wait for
1 from
below

&& has_seq1(rcvpkt)

rdt_rcv(rcvpkt) &&
not corrupt(rcvpkt) &&
has_seq0(rcvpkt)
udt_send(sndpkt)

deliver_data(data)
udt_send(sndpkt)

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”
whether “expected”
pkt should have seq #
of 0 or 1

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
ACK/NAK received
OK at sender


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


rdt2.2: sender, receiver fragments
rdt_send(data)
udt_send(sndpkt)
rdt_rcv(rcvpkt) &&
Wait for
call 0 from
above

rdt_rcv(rcvpkt) &&
(corrupt(rcvpkt) ||
has_seq1(rcvpkt))
udt_send(sndpkt)

Wait for
0 from
below

sender FSM
fragment

isACK(rcvpkt,1) )
udt_send(sndpkt)

Wait for
ACK
0

rdt_rcv(rcvpkt)
&& isACK(rcvpkt,0)

receiver FSM
fragment

Λ

&& has_seq1(rcvpkt)
deliver_data(data)
sndpkt = make_pkt(ACK1, chksum)
udt_send(sndpkt)


rdt3.0: channels with errors and loss
new assumption:
underlying channel can
also lose packets
(data, 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 seq. #’s
already handles this
 receiver must specify seq
# of pkt being ACKed
requires countdown timer

rdt3.0 sender
rdt_send(data)
udt_send(sndpkt)
start_timer

rdt_rcv(rcvpkt)

Λ

rdt_rcv(rcvpkt)
&& isACK(rcvpkt,1)

rdt_rcv(rcvpkt) &&
isACK(rcvpkt,0) )

timeout
udt_send(sndpkt)
start_timer
rdt_rcv(rcvpkt)
&& isACK(rcvpkt,0)
stop_timer

stop_timer
timeout
udt_send(sndpkt)
start_timer

Λ

Wait
for
ACK0

Wait for
call 0from
above

Λ

rdt_rcv(rcvpkt) &&
isACK(rcvpkt,1) )

Wait
for
ACK1

Wait for
call 1 from
above
rdt_send(data)

rdt_rcv(rcvpkt)

Λ

udt_send(sndpkt)
start_timer


rdt3.0 in action
receiver

sender
send pkt0
rcv ack0
send pkt1
rcv ack1
send pkt0

pkt0
ack0
pkt1
ack1
pkt0
ack0

(a) no loss

rcv pkt0
send ack0
rcv pkt1
send ack1
rcv pkt0
send ack0

receiver

sender
send pkt0
rcv ack0
send pkt1

pkt0
ack0

rcv pkt0
send ack0

pkt1

X

loss

timeout
resend pkt1
rcv ack1
send pkt0

pkt1
ack1
pkt0
ack0

rcv pkt1
send ack1
rcv pkt0
send ack0

(b) packet loss

rdt3.0 in action
receiver

sender
send pkt0
rcv ack0
send pkt1

pkt0
ack0
pkt1
ack1

X

rcv pkt0
send ack0
rcv pkt1
send ack1

loss

timeout
resend pkt1
rcv ack1
send pkt0

pkt1
ack1
pkt0
ack0

(c) ACK loss

rcv pkt1

(detect duplicate)

send ack1

rcv pkt0
send ack0

receiver

sender
send pkt0
rcv ack0
send pkt1

pkt0

rcv pkt0
send ack0

ack0
pkt1

rcv pkt1
send ack1

ack1

timeout
resend pkt1
rcv ack1
send pkt0
rcv ack1
send pkt0

pkt1

rcv pkt1

pkt0
ack1
ack0
pkt0

(detect duplicate)

ack0

(detect duplicate)

send ack1
rcv pkt0
send ack0
rcv pkt0
send ack0

(d) premature timeout/ delayed ACK

Performance of rdt3.0



rdt3.0 is correct, but performance stinks
e.g.: 1 Gbps link, 15 ms prop. delay, 8000 bit packet:
8000 bits
L
Dtrans = R =
109 bits/sec

= 8 microsecs

 U sender: utilization – fraction of time sender busy sending
U

sender =

L/R
RTT + L / R

=

.008
30.008

= 0.00027

 if RTT=30 msec, 1KB pkt every 30 msec: 33kB/sec thruput
over 1 Gbps link


network protocol limits use of physical resources!

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
last packet bit arrives, send
ACK

RTT

ACK arrives, send next
packet, t = RTT + L / R

U

sender =

L/R
RTT + L / R

=

.008
30.008

= 0.00027


Pipelined protocols
pipelining: sender allows multiple, “in-flight”, yetto-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

Pipelining: increased utilization
sender

receiver

first packet bit transmitted, t = 0
last bit transmitted, t = L / R
first packet bit arrives
last packet bit arrives, send ACK
last bit of 2nd packet arrives, send ACK
last bit of 3rd packet arrives, send ACK

RTT

ACK arrives, send next
packet, t = RTT + L / R

3-packet pipelining increases
utilization by a factor of 3!

U

sender =

3L / R
RTT + L / R

=

.0024
30.008

= 0.00081


Pipelined protocols: overview
Go-back-N:
 sender can have up to
N unacked packets in
pipeline
 receiver only sends
cumulative ack

Selective Repeat:
 sender can have up to N
unack’ed packets in
pipeline
 rcvr sends individual ack
for each packet

 doesn’t ack packet if
there’s a gap



sender has timer for
oldest unacked packet
 when timer expires,
retransmit all unacked
packets



sender maintains timer
for each unacked packet
 when timer expires,
retransmit only that
unacked packet

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

GBN: sender extended FSM
rdt_send(data)

Λ
base=1
nextseqnum=1

rdt_rcv(rcvpkt)
&& corrupt(rcvpkt)

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)
timeout
start_timer
Wait
udt_send(sndpkt[base])
udt_send(sndpkt[base+1])
…
udt_send(sndpkt[nextseqnum1])
rdt_rcv(rcvpkt) &&
notcorrupt(rcvpkt)
base = getacknum(rcvpkt)+1
If (base == nextseqnum)
stop_timer
else
start_timer


GBN: receiver extended FSM
default
udt_send(sndpkt)

Λ

Wait
expectedseqnum=1
sndpkt =
make_pkt(expectedseqnum,ACK,chksum)

rdt_rcv(rcvpkt)
&& notcurrupt(rcvpkt)
&& hasseqnum(rcvpkt,expectedseqnum)
deliver_data(data)
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 #

GBN in action
sender window (N=4)
012345678
012345678
012345678
012345678

012345678
012345678

sender
send
send
send
send

pkt0
pkt1
pkt2
pkt3
(wait)

rcv ack0, send pkt4
rcv ack1, send pkt5
ignore duplicate ACK

012345678
012345678
012345678
012345678

pkt 2 timeout
send pkt2
send pkt3
send pkt4
send pkt5

receiver

Xloss

receive pkt0, send ack0
receive pkt3, discard,
(re)send ack1
(re)send ack1
(re)send ack1
rcv pkt2, deliver, send ack2

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


Selective repeat: sender, receiver windows


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:


ignore


Selective repeat in action
sender window (N=4)
012345678
012345678
012345678
012345678

012345678
012345678

sender
send
send
send
send

pkt0
pkt1
pkt2
pkt3
(wait)

receiver

Xloss

rcv ack0, send pkt4
rcv ack1, send pkt5
record ack3 arrived

012345678
012345678
012345678
012345678

pkt 2 timeout
send pkt2
record ack4 arrived
record ack4 arrived

receive pkt3, buffer,
send ack3
send ack4
send ack5
rcv pkt2; deliver pkt2,
pkt3, pkt4, pkt5; send ack2

Q: what happens when ack2 arrives?

Selective repeat:
dilemma
example:






seq #’s: 0, 1, 2, 3
window size=3
receiver sees no
difference in two
scenarios!
duplicate data
accepted as new in
(b)

receiver window
(after receipt)

sender window
(after receipt)
0123012

pkt0

0123012

pkt1

0123012

0123012

pkt2

0123012

0123012

pkt3

0123012

pkt0

(a) no problem

0123012

X
will accept packet
with seq number 0

receiver can’t see sender side.
receiver behavior identical in both cases!
something’s (very) wrong!
pkt0

0123012

pkt1

0123012

0123012

Q: what relationship
between seq # size
and window size to
avoid problem in (b)?

0123012

pkt2

0123012

X
X
timeout
retransmit pkt0 X
0123012

(b) oops!

pkt0

0123012

will accept packet
with seq number 0

Chapter 3 outline
3.1 transport-layer
services
demultiplexing
3.3 connectionless
transport: UDP
data transfer

transport: TCP
 flow control

control


TCP: Overview


RFCs: 793,1122,1323, 2018, 2581

point-to-point:



 one sender, one receiver




 bi-directional data flow
in same connection
 MSS: maximum segment
size

reliable, in-order byte
steam:
 no “message
boundaries”



connection-oriented:
 handshaking (exchange
of control msgs) inits
sender, receiver state
before data exchange

pipelined:
 TCP congestion and
flow control set window
size

full duplex data:



flow controlled:
 sender will not
overwhelm receiver

TCP segment structure
32 bits
URG: urgent data
(generally not used)
ACK: ACK #
valid
PSH: push data now
(generally not used)
RST, SYN, FIN:
connection estab
(setup, teardown
commands)
Internet
checksum
(as in UDP)

source port #

dest port #

sequence number
acknowledgement number
head not
UAP R S F
len used

checksum

receive window
Urg data pointer

options (variable length)

counting
by bytes
of data
(not segments!)
# bytes
rcvr willing
to accept

application
data
(variable length)


TCP seq. numbers, ACKs
sequence numbers:
 byte stream “number” of
first byte in segment’s
data
acknowledgements:
 seq # of next byte
expected from other side
 cumulative ACK
Q: how receiver handles
out-of-order segments
 A: TCP spec doesn’t say,
- up to implementor

outgoing segment from sender
source port #

dest port #

sequence number
rwnd
checksum

urg pointer

window size
N

sender sequence number space
sent
ACKed

sent, not- usable not
yet ACKed but not usable
(“in-flight”) yet sent

incoming segment to sender
source port #

dest port #

sequence number
rwnd
A
checksum

urg pointer


TCP seq. numbers, ACKs
Host B

Host A

User
types
‘C’

host ACKs
receipt
of echoed
‘C’

Seq=42, ACK=79, data = ‘C’

Seq=79, ACK=43, data = ‘C’

host ACKs
receipt of
‘C’, echoes
back ‘C’

Seq=43, ACK=80

simple telnet scenario


TCP round trip time, timeout
Q: how to set TCP
timeout value?


Q: how to estimate RTT?


longer than RTT
 but RTT varies





too short: premature
timeout, unnecessary
retransmissions
too long: slow reaction
to segment loss



SampleRTT: measured
time from segment
transmission until ACK
receipt
 ignore retransmissions
SampleRTT will vary, want
estimated RTT “smoother”
 average several recent
measurements, not just
current SampleRTT


EstimatedRTT = (1- α)*EstimatedRTT + α*SampleRTT



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

350

RTT: gaia.cs.umass.edu to fantasia.eurecom.fr
300

RTT (milliseconds)



exponential weighted moving average
influence of past sample decreases exponentially fast
typical value: α = 0.125

s dnoc eslli m TT R
i
(



250

200

sampleRTT

150

EstimatedRTT

100
1

8

15

22

29

36

43

50

57

64

time (seconds)

71

time (seconnds)

SampleRTT

Estimated RTT

78

85

92

99

106




timeout interval: EstimatedRTT plus “safety margin”
 large variation in EstimatedRTT -> larger safety margin



estimate SampleRTT deviation from EstimatedRTT:
DevRTT = (1-β)*DevRTT +
β*|SampleRTT-EstimatedRTT|
(typically, β =
0.25)

TimeoutInterval = EstimatedRTT + 4*DevRTT
estimated RTT

“safety margin”


Chapter 3 outline
3.1 transport-layer
services
demultiplexing
3.3 connectionless
transport: UDP
data transfer

transport: TCP
 flow control

control


TCP reliable data transfer


TCP creates rdt service
on top of IP’s unreliable
service
 pipelined segments
 cumulative acks
 single retransmission
timer



retransmissions
triggered by:

let’s initially consider
simplified TCP sender:
 ignore duplicate acks
 ignore flow control,
congestion control

 timeout events
 duplicate acks


TCP sender events:
data rcvd from app:
 create segment with
seq #
 seq # is byte-stream
number of first data
byte in segment
 start timer if not
already running
 think of timer as for
oldest unacked
segment
 expiration interval:
TimeOutInterval

timeout:
 retransmit segment
that caused timeout
 restart timer
ack rcvd:
 if ack acknowledges
previously unacked
segments
 update what is known
to be ACKed
 start timer if there are
still unacked segments

TCP sender (simplified)
data received from application above

Λ
NextSeqNum = InitialSeqNum
SendBase = InitialSeqNum

wait
for
event

create segment, seq. #: NextSeqNum
pass segment to IP (i.e., “send”)
NextSeqNum = NextSeqNum + length(data)
if (timer currently not running)
start timer
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
}


TCP: retransmission scenarios
Host B

Host A

Host B

Host A

SendBase=92

X

Seq=92, 8 bytes of data

ACK=100


ACK=100

t uoe m
it

t uoe m
it


SendBase=100

ACK=100
ACK=120
Seq=92, 8
bytes of data

SendBase=120
ACK=120
SendBase=120

lost ACK scenario

premature timeout

TCP: retransmission scenarios
Host B

Host A


t uoe m
it

X

ACK=100

ACK=120


cumulative ACK

TCP ACK generation

[RFC 1122, RFC 2581]

event at receiver

TCP receiver action

arrival of in-order segment with
expected seq #. All data up to
expected seq # already ACKed

delayed ACK. Wait up to 500ms
for next segment. If no next segment,
send ACK

arrival of in-order segment with
expected seq #. One other
segment has ACK pending

immediately send single cumulative
ACK, ACKing both in-order segments

arrival of out-of-order segment
higher-than-expect seq. # .
Gap detected

immediately send duplicate ACK,
indicating seq. # of next expected byte

arrival of segment that
partially or completely fills gap

immediate send ACK, provided that
segment starts at lower end of gap

TCP fast retransmit


time-out period often
relatively long:
 long delay before
resending lost packet



detect lost segments
via duplicate ACKs.
 sender often sends
many segments backto-back
 if segment is lost, there
will likely be many
duplicate ACKs.

TCP fast retransmit

if sender receives 3
ACKs for same data
(“triple duplicate ACKs”),
(“triple duplicate ACKs”),

resend unacked
segment with smallest
seq #

 likely that unacked
segment lost, so don’t
wait for timeout


TCP fast retransmit
Host B

Host A


X
ACK=100

t uoe m
it

ACK=100
ACK=100
ACK=100

fast retransmit after sender
receipt of triple duplicate ACK


Chapter 3 outline
3.1 transport-layer
services
demultiplexing
3.3 connectionless
transport: UDP
data transfer

transport: TCP
 flow control

control


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

application
process
application

TCP
code

IP
code

flow control

receiver controls sender, so
sender won’t overflow
receiver’s buffer by transmitting
too much, too fast

OS

TCP socket
receiver buffers

from sender

receiver protocol stack

TCP flow control


receiver “advertises” free
buffer space by including
rwnd value in TCP header
of receiver-to-sender
segments
 RcvBuffer size set via
socket options (typical default
is 4096 bytes)
 many operating systems
autoadjust RcvBuffer





sender limits amount of
unacked (“in-flight”) data to
receiver’s rwnd value
guarantees receive buffer
will not overflow

to application process

RcvBuffer
rwnd

buffered data
free buffer space

TCP segment payloads

receiver-side buffering


Chapter 3 outline
3.1 transport-layer
services
demultiplexing
3.3 connectionless
transport: UDP
data transfer

transport: TCP
 flow control

control


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

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

network

network

Socket clientSocket =
newSocket("hostname","port
number");

Socket connectionSocket =
welcomeSocket.accept();

Agreeing to establish a connection
2-way handshake:

Let’s talk
ESTAB

OK

Q: will 2-way handshake
always work in
network?
ESTAB






choose x
ESTAB

req_conn(x)
acc_conn(x)



variable delays
retransmitted messages
(e.g. req_conn(x)) due to
message loss
message reordering
can’t “see” other side

ESTAB


Agreeing to establish a connection
2-way handshake failure scenarios:
choose x
retransmit
req_conn(x)

req_conn(x)

choose x
ESTAB

acc_conn(x)

retransmit
req_conn(x)
ESTAB

ESTAB
req_conn(x)

client
terminates

connection
x completes

retransmit
data(x+1)
server
forgets x
ESTAB

half open connection!
(no client!)

client
terminates

req_conn(x)

ESTAB

acc_conn(x)
data(x+1)
connection
x completes

req_conn(x)
data(x+1)

accept
data(x+1)
server
forgets x
ESTAB
accept
data(x+1)


TCP 3-way handshake
client state
LISTEN
SYNSENT

server state
LISTEN

choose init seq num, x
send TCP SYN msg

received SYNACK(x)
indicates server is live;
ESTAB
send ACK for SYNACK;
this segment may contain
client-to-server data

SYNbit=1, Seq=x

choose init seq num, y
send TCP SYNACK
SYN RCVD
msg, acking SYN

SYNbit=1, Seq=y
ACKbit=1; ACKnum=x+1

ACKbit=1, ACKnum=y+1
received ACK(y)
indicates client is live

ESTAB


TCP 3-way handshake: FSM
closed
Socket connectionSocket =
welcomeSocket.accept();

Λ
SYN(x)
SYNACK(seq=y,ACKnum=x+1)
create new socket for
communication back to client

listen

SYN(seq=x)

SYN
sent

SYN
rcvd

ACK(ACKnum=y+1)

Socket clientSocket =
newSocket("hostname","port
number");

ESTAB

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

Λ

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


TCP: closing a connection
client state

server state

ESTAB

ESTAB
clientSocket.close()

FIN_WAIT_1

FIN_WAIT_2

can no longer
send but can
receive data

FINbit=1, seq=x
ACKbit=1; ACKnum=x+1

wait for server
close

FINbit=1, seq=y

TIMED_WAIT
timed wait
for 2*max
segment lifetime

ACKbit=1; ACKnum=y+1

CLOSE_WAIT
can still
send data

LAST_ACK
can no longer
send data

CLOSED

CLOSED

Chapter 3 outline
3.1 transport-layer
services
demultiplexing
3.3 connectionless
transport: UDP
data transfer

transport: TCP
 flow control

control


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!


Causes/costs of congestion: scenario 1




λout

Host A

unlimited shared
output link buffers

Host B

R/2

delay



two senders, two
receivers
one router, infinite
buffers
output link capacity: R
no retransmission

throughput:

λout



original data: λin



λin R/2
maximum per-connection
throughput: R/2



λin R/2
large delays as arrival rate,
λin, approaches capacity




one router, finite buffers
sender retransmission of timed-out packet
 application-layer input = application-layer output: λin = λout
 transport-layer input includes retransmissions : λin λin
‘

λin : original data
λ'in: original data, plus

λout

retransmitted data
Host A

Host B

finite shared output
link buffers

λout

idealization: perfect
knowledge
 sender sends only when
router buffers available

R/2


copy

λin

R/2

λout

retransmitted data
A

Host B

free buffer space!

link buffers

Idealization: known loss



packets can be lost,
dropped at router due
to full buffers
sender only resends if
packet known to be lost

copy

λout

retransmitted data
A

no buffer space!

Host B




packets can be lost,
dropped at router due
to full buffers
sender only resends if
packet known to be lost

R/2
when sending at R/2,
some packets are
retransmissions but
asymptotic goodput
is still R/2 (why?)

λout

Idealization: known loss


λin

R/2

λout

retransmitted data
A

free buffer space!

Host B





packets can be lost, dropped
at router due to full buffers
sender times out prematurely,
sending two copies, both of
which are delivered

R/2

λin
λ'in

timeout
copy

A

some packets are
retransmissions
including duplicated
that are delivered!

λout

Realistic: duplicates

λin

R/2

λout

free buffer space!

Host B





packets can be lost, dropped
at router due to full buffers
sender times out prematurely,
sending two copies, both of
which are delivered

R/2
some packets are
retransmissions
including duplicated
that are delivered!

λout

Realistic: duplicates

λin

R/2

“costs” of congestion:



more work (retrans) for given “goodput”
unneeded retransmissions: link carries multiple copies of pkt
 decreasing goodput






four senders
multihop paths
timeout/retransmit
Host A

Q: what happens as λin and λin’

increase ?
A: as red λin’ increases, all arriving
blue pkts at upper queue are
dropped, blue throughput  0


λout

Host B

retransmitted data
link buffers

Host D
Host C



λout

C/2

λin’

C/2

another “cost” of congestion:
 when packet dropped, any “upstream
transmission capacity used for that packet was
wasted!

Approaches towards congestion control
two broad approaches towards congestion control:
end-end congestion
control:





no explicit feedback
from network
congestion inferred
from end-system
observed loss, delay
approach taken by
TCP

network-assisted
congestion control:


routers provide
feedback to end systems
 single bit indicating
congestion (SNA,
DECbit, TCP/IP ECN,
ATM)
 explicit rate for
sender to send at

Case study: ATM ABR congestion control
ABR: available bit rate:





“elastic service”
if sender’s path
“underloaded”:
 sender should use
available bandwidth
if sender’s path
congested:
 sender throttled to
minimum guaranteed
rate

RM (resource management)
cells:






sent by sender, interspersed
with data cells
bits in RM cell set by
switches (“network-assisted”)
 NI bit: no increase in rate
(mild congestion)
 CI bit: congestion
indication
RM cells returned to sender
by receiver, with bits intact

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

Chapter 3 outline
3.1 transport-layer
services
demultiplexing
3.3 connectionless
transport: UDP
data transfer

transport: TCP
 flow control

control


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 loss

AIMD saw tooth
behavior: probing
for bandwidth

cwnd: TCP sender
congestion window size



additively increase window size …
…. until loss occurs (then cut window in half)

time

TCP Congestion Control: details
sender sequence number space
cwnd

last byte
ACKed



sent, notyet ACKed
(“in-flight”)

last byte
sent

sender limits transmission:

TCP sending rate:
 roughly: send cwnd
bytes, wait RTT for
ACKS, then send
more bytes
rate

~
~

cwnd
RTT

bytes/sec

LastByteSent< cwnd
LastByteAcked



cwnd is dynamic, function
of perceived network
congestion

TCP Slow Start
when connection begins,
increase rate
exponentially until first
loss event:
 initially cwnd = 1 MSS
 double cwnd every RTT
 done by incrementing
cwnd for every ACK
received


summary: initial rate is
slow but ramps up
exponentially fast

RTT



Host B

Host A

one segm
ent

two segm
ents

four segm

ents

time


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)


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

Summary: TCP Congestion Control
duplicate ACK
dupACKcount++
Λ
cwnd = 1 MSS
ssthresh = 64 KB
dupACKcount = 0

slow
start

timeout
ssthresh = cwnd/2
cwnd = 1 MSS
dupACKcount = 0
retransmit missing segment

dupACKcount == 3
ssthresh= cwnd/2
cwnd = ssthresh + 3

New
ACK!
new ACK
cwnd = cwnd+MSS
dupACKcount = 0
transmit new segment(s), as allowed
cwnd > ssthresh
Λ
timeout
ssthresh = cwnd/2
cwnd = 1 MSS
dupACKcount = 0

timeout
ssthresh = cwnd/2
cwnd = 1
dupACKcount = 0

.

New
ACK!

new ACK
cwnd = cwnd + MSS (MSS/cwnd)
dupACKcount = 0

congestion
avoidance
duplicate ACK
dupACKcount++

New
ACK!
New ACK
cwnd = ssthresh
dupACKcount = 0

fast
recovery

dupACKcount == 3
ssthresh= cwnd/2
cwnd = ssthresh + 3

duplicate ACK
cwnd = cwnd + MSS


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
avg TCP thruput =

3 W
bytes/sec
4 RTT

W

W/2


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

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

TCP connection 2

bottleneck
router
capacity R


Why is TCP fair?
two competing sessions:


additive increase gives slope of 1, as throughout increases
multiplicative decrease decreases throughput proportionally
R
Connection 2 throughput



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

Connection 1 throughput

R

Fairness (more)
Fairness and UDP
 multimedia apps often
do not use TCP



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

Chapter 3: summary




principles behind
transport layer services:
 multiplexing,
demultiplexing
 flow control
instantiation,
implementation in the
Internet

next:
 leaving the
network “edge”
(application,
transport layers)
 into the network
“core”

 UDP
 TCP

Transport layer (computer networks)

More Related Content

What's hot (20)

Similar to Transport layer (computer networks) (20)

Recently uploaded (20)

Transport layer (computer networks)

Editor's Notes