transport layer 3-1 ece 5650 tl: reliable data transfer

Transport Layer 3-1

ECE 5650 TL: Reliable Data Transfer

Transport Layer 3-2

Recap: transport-layer protocols IP service model

Host-to-host best-effort delivery service, datagram

Unreliable serivce UDP: unreliable, unordered

delivery Proc to proc segment delivery: no-

frills extension of “best-effort” IP (Multiplexing /demultiplexing)

Error checking TCP:

reliable, in-order delivery: flow control, sequence number, ack, timer, etc

congestion control to prevent TCP connection from swamping the links and switches

services not available in either TCP or UDP: delay guarantees bandwidth guarantees

application

transportnetworkdata linkphysical

application

transportnetworkdata linkphysical

networkdata linkphysical



networkdata linkphysicalnetwork

data linkphysical

logical end-end transport

Transport Layer 3-3

Recap: Multiplexing/demultiplexing (encode/decode)

application

transport

network

link

physical

P1 application

transport

network

link

physical

application

transport

network

link

physical

P2P3 P4P1

host 1 host 2 host 3

= process= socket

The job of delivering received transport-layer segments to correct socket.

Each socket has a unique identifier (a set offields in each segment)

Demultiplexing at rcv host:

The job of gathering data from differentsockets, enveloping data with header (later used for demultiplexing) and passing the data to the network layer.

Multiplexing at send host:

Transport Layer 3-4

Recap: Connectionless demux

DatagramSocket serverSocket = new DatagramSocket(6428);

ClientIP:B

P2

client IP: A

P1P1P3

serverIP: C

SP: 6428

DP: 9157

SP: 9157

DP: 6428

SP: 6428

DP: 5775

SP: 5775

DP: 6428

SP = Source Port, DP = Destination PortSource Packet provides “return address” that destination host can use to reply

port 9157 port 6428 port 5775

One Server Socket

Transport Layer 3-5

Recap: Conn-oriented demux

ClientIP:B

P1

client IP: A

P1P2P4

serverIP: C

SP: 9157

DP: 80

SP: 9157

DP: 80

P5 P6 P3

D-IP:CS-IP: A

D-IP:C

S-IP: B

SP: 5775

DP: 80

D-IP:CS-IP: B

SP = Source Port, DP = Destination Port, S-IP = Source IP, D-IP = Destination IP

Three Server Sockets

Transport Layer 3-6

Recap: UDP Segment Structure

UDP segment has an 8 byte header.

Length of UDP segment, in bytes, including header.

Minimum length is 8 and maximum is 64k.

source port # dest port #

32 bits

Applicationdata

(message)

UDP segment format

length (16-bit) checksum (16-bit)

Transport Layer 3-7

Recap: Internet Checksum: Sender Example Note

When adding numbers, a carryout from the most significant bit needs to be added to the result

Segment has two 16-bit integers and the checksum

Add the two 16-bit integers

compute the checksum by taking the 1’s complement of

the sum.

1 1 1 1 0 0 1 1 0 0 1 1 0 0 1 1 01 1 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1

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

1 1 0 1 1 1 0 1 1 1 0 1 1 1 1 0 01 0 1 0 0 0 1 0 0 0 1 0 0 0 0 1 1

wraparound:add carry to LSB

sumchecksum

Transport Layer 3-8

TL 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 reliable data transfer flow control connection

management

3.6 Principles of congestion control

3.7 TCP congestion control

Transport Layer 3-9

Principles of Reliable data transfer

Reliable transfer: no data corrupted or loss, and in-order delivery important in app., 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 3-10

Reliable data transfer: getting started

sendside

receiveside

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

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

receiver

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

deliver_data(): called by rdt to deliver data to

upper


Reliable data transfer: getting startedWe’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

state1

state2

event causing state transition (when..happens)actions taken on state transition (then this happens)

state: when in this “state” next state

uniquely determined by

next event

eventactions

Symbol above or below the line denote the lack of an action or event


Finite State Machine

FSM is a model of behavior composed of states, transitions and events State stores info about the past,

reflecting the system start to the present moment

Transition indicates a state change, triggered by an event

Events: activity to be performed at a given moment: entry, exit, input

FSM often represented using a state diagram


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 call from above packet = make_pkt(data)

udt_send(packet)

rdt_send(data)

extract (packet,data)deliver_data(data)

Wait for call from

below

rdt_rcv(packet)

sender receiver


Rdt2.0: channel with bit errors but does not lose packets

underlying channel may flip bits in packet Solution: 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 “please repeat that”. 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


Automatic Repeat reQuest protocols (ARQ)

Protocols based on ACKs and NAKs. ARQ Protocols have 3 capabilities:

Error detection of bit errors which requires additional bits added to the transmitted packet.

Receiver feedback: ACK and NAK bit (1=ACK, 0=NAK) to be added to the transmitted packet.

Retransmission: a packet with errors will be retrasmitted.


rdt2.0: FSM specification

Wait for call from above

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

extract(rcvpkt,data)deliver_data(data)udt_send(ACK)

rdt_rcv(rcvpkt) && notcorrupt(rcvpkt)

rdt_rcv(rcvpkt) && isACK(rcvpkt)

udt_send(sndpkt)

rdt_rcv(rcvpkt) && isNAK(rcvpkt)

udt_send(NAK)

rdt_rcv(rcvpkt) && corrupt(rcvpkt)

Wait for ACK or

NAK

Wait for call from

belowsender

receiverrdt_send(data)

Sender can not send another packet until receiver sends back ACK/NAK. (send-and-wait protocols)


rdt2.0: operation with no errors


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




udt_send(sndpkt)


udt_send(NAK)


Wait for ACK or

NAK

Wait for call from

below

rdt_send(data)


rdt2.0: error scenario


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




udt_send(sndpkt)


udt_send(NAK)


Wait for ACK or

NAK

Wait for call from

below

rdt_send(data)


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

packet if ACK/NAK garbled/corrupt

sender adds sequence number to each packet, 1-bit that changes if a new packet is sent

receiver discards (doesn’t deliver up) duplicate pkt

ACK/NAK packets do not need the sequence number of the packet acknowledged since sender knows it is for the last packet (we assume no packet loss)


rdt2.1: sender, handles garbled ACK/NAKs

Wait for call 0 from

above

sndpkt = make_pkt(0, data, checksum)udt_send(sndpkt)

rdt_send(data)

Wait for ACK or NAK 0 udt_send(sndpkt)

rdt_rcv(rcvpkt) && ( corrupt(rcvpkt) ||isNAK(rcvpkt) )


rdt_send(data)

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

udt_send(sndpkt)

rdt_rcv(rcvpkt) && ( corrupt(rcvpkt) ||isNAK(rcvpkt) )

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


above

Wait for ACK or NAK 1


rdt2.1: receiver, handles garbled ACK/NAKs

Wait for 0 from below

sndpkt = make_pkt(NAK, chksum)udt_send(sndpkt)

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

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

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



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

rdt_rcv(rcvpkt) && (corrupt(rcvpkt)

sndpkt = make_pkt(ACK, chksum)udt_send(sndpkt)

rdt_rcv(rcvpkt) && not corrupt(rcvpkt) && has_seq1(rcvpkt)

rdt_rcv(rcvpkt) && (corrupt(rcvpkt))

sndpkt = make_pkt(ACK, chksum)udt_send(sndpkt)

sndpkt = make_pkt(NAK, chksum)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 “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 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


above


rdt_send(data)

udt_send(sndpkt)

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

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

Wait for ACK

0

sender FSMfragment



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

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

udt_send(sndpkt)

receiver FSMfragment


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 # of pkt being ACKed

requires countdown timer

timeout is worst-case time (taking all types of delays into account) for packet to reach receiver and for an ACK to be received


rdt3.0 sender

sndpkt = make_pkt(0, data, checksum)udt_send(sndpkt)start_timer

rdt_send(data)

Wait for

ACK0

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


above

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

rdt_send(data)


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


stop_timerstop_timer

udt_send(sndpkt)start_timer

timeout

udt_send(sndpkt)start_timer

timeout

rdt_rcv(rcvpkt)

Wait for call 0from

above

Wait for

ACK1

rdt_rcv(rcvpkt)


rdt3.0 in action

Alternating-bit protocol because packet sequence numbers alternate between 0 and 1


rdt3.0 in action


Performance of rdt3.0 rdt3.0 works, but performance stinks example: 1 Gbps link, 15 ms e-e propagation delay, 1KByte

packet:

Ttransmit

= 8kb/pkt10^9 b/sec

= 8 microsec

U sender: utilization – fraction of time sender busy sending 1 KByte packet every 30 msec -> 267 kbps throughput

over 1 Gbps link network protocol limits use of physical resources!

U sender

= .008

30.008 = 0.00027

microseconds

L / R

RTT + L / R =

L (packet length in bits)R (transmission rate, bps)

=

last bit arrives at receiver at 15 ms + 8 microsec = 15.008 msACK comes back at sender at 15.008 ms + 15 ms = 30.008 ms (assuming ACK size is small)


rdt3.0: stop-and-wait operation

first packet bit transmitted, t = 0

sender receiver

RTT

last packet bit transmitted, t = L / R

first packet bit arriveslast packet bit arrives, send ACK

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

U sender

= .008

30.008 = 0.00027

microseconds

L / R

RTT + L / R =


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


Pipelining: increased utilization

first packet bit transmitted, t = 0

sender receiver

RTT

last bit transmitted, t = L / R

first packet bit arriveslast packet bit arrives, send ACK

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

last bit of 2nd packet arrives, send ACKlast bit of 3rd packet arrives, send ACK

U sender

= .024

30.008 = 0.0008

microseconds

3 * L / R

RTT + L / R =

Increase utilizationby a factor of 3!


Go-Back-N (Sliding Window)Sender: There is a k-bit sequence # in packet header “window” of up to N, consecutive unacknowledged sent/can-be-sent packets allowed window moves by 1 packet at a time when its 1st sent pkt is acknowledged (standard behavior)

Sender must respond to three types of events: 1- Invocation from above: application layers tries to send a packet, if window is full

then packet is returned otherwise the packet is accepted and sent. 2- Receipt of an ACK: One ACK(n) received indicates that all pkts up to, including seq

# n have been received - “cumulative ACK” may receive duplicate ACKs (when receiver receives out-of-order packets)

3- A timeout event (only cause of retransmission): timer for each in-flight pkt. if timeout occurs: retransmit packets that have not been acknowledged.

window cannot contain acknowledged pkts


GBN: sender extended FSM

Wait start_timerudt_send(sndpkt[base])udt_send(sndpkt[base+1])…udt_send(sndpkt[nextseqnum-1])

timeout

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 = getacknum(rcvpkt)+1If (base == nextseqnum) stop_timer else start_timer


base=1nextseqnum=1



GBN: receiver extended FSM

ACK-only: always send ACK for correctly-received pkt with highest in-order seq # need only remember expectedseqnum

When out-of-order pkt is received: discard (don’t buffer) -> no receiver buffering! Re-ACK pkt with highest in-order sequence #

Wait

udt_send(sndpkt)

default

rdt_rcv(rcvpkt) && notcurrupt(rcvpkt) && hasseqnum(rcvpkt,expectedseqnum)

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

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


GBN inaction

Window-size = 4


Notes on GBN: Receiver discards out-of-order packets and does

not even buffer them for future use: Receiver must deliver data in order to application layer If out-of-order packets are buffered and in-order

packets are lost then a retransmission will re-send the out-of-order packets.

Simplicity of receiver buffering due to discarding One disadvantage of not buffering is that a

retransmission could be also lost which would require additional retransmissions.

A single packet error can cause GBN sender to retransmit larger #pkts as the window size increases.


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


Selective repeat: sender, receiver windows

Window may contain acknowledged pkts (unlike GBN)


Selective repeat

data from above : if next available seq # in

window, send pkt

timeout(n): resend pkt n, restart

timer

ACK(n) in window:

mark pkt n as received if n smallest unACKed

pkt, advance window base to next unACKed seq #

pkt n in window

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 previously acknowledged

ACK(n), necessary to move sender’s window

otherwise: ignore

receiver eventssender events


Selective repeat in action


Selective repeat: too large window 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)

To avoid this condition:the sequence number space (k) must be at least twice as large as the window size (w) or k >= 2w


Reliable Data Transfer Mechanisms Checksum: used to detect bit errors in transmitted pkt. Timer: used to retransmit a packet when the packet or its ACK

was lost. Duplicate copies of pkt could be received due to premature timeouts when the packet or its ACK are delayed.

Sequence number: used for sequential numbering of packets sent. Gaps in seq #s indicate lost/delayed packts while duplicate seq #s indicate duplicate pkts.

Acknowledgement (ACK): used by receiver to tell sender a packet (individual ACK) or packets (cumulative) have been received.

Negative Acknowledgement (NAK): used by receiver to tell sender a packet has not been received correctly.

Pipelining: used to allow sending packets even if prior ones were not acknowledged and hence increasing the utilization of the sender.

Window: used to restrict sender to send packets with seq #s within a range (window size) without waiting for an acknowledgement.


Summary

Automatic Repeat reQuest protocols (ARQ) Send-and-wait protocols The use of Sequences numbers in reliable data

transfers Low Utilization of the Send-and-wait protocols 2 approaches to Pipelined error recovery:

Go-Back-N (GBN)/sliding-window protocol Selective Repeat (SR)

transport layer 3-1 ece 5650 tl: reliable data transfer

Documents