datacommchapter 6 part 2


2/57

2

The Internet model has three protocols at theThe Internet model has three protocols at the

transport layer: UDP, TCP & SCTPtransport layer: UDP, TCP & SCTP

The UDP is the simplest of the three.The UDP is the simplest of the three.

The TCP is a complex transport layer protocol.The TCP is a complex transport layer protocol.

The SCTP is the new transport layer protocolThe SCTP is the new transport layer protocol

that is designed for that is designed formultihomedmultihomed,, multistreammultistreamapplications such as multimedia.applications such as multimedia.


3/57

3

Position of UDP, TCP, and SCTP in TCP/IP suite


4/57

4

6.4: USER DATAGRAM PROTOCOL (UDP)6.4: USER DATAGRAM PROTOCOL (UDP)

The User Datagram Protocol (UDP) is called aThe User Datagram Protocol (UDP) is called a connectionless, unreliable transport protocol. It connectionless, unreliable transport protocol. It

does not add anything to the services of IP does not add anything to the services of IP

except to provide processexcept to provide process--toto--processprocess communication instead of host communication instead of host--toto--hosthost

communication.communication.


5/57

5

UDP is a simpler protocol compared to TCP.

It has no connection, flow control, congestion control, sequencing,

packet acknowledgement and reliability. There is no control mechanism for detecting lost packets, out-of-order

packets and duplicate packets.

It only provides error control for detecting corrupted packets using

checksum. Processes which involved sending small message or do not care about

reliability can use UDP.

It is used by application protocols such as TFTP, SNMP, RTP and etc.

UDP is also a suitable transport protocol for multicasting of real timemultimedia.

In transport layer of TCP/IP model.


6/57

6

Well-known ports used with UDP

Some ports number can be used by both UDP and TCP.


7/57

7

User datagram format

Source port number: the port number used by the process running on the source host.

Destination port number: the port number used by the process running on the

destinationCheck sum: detects errors over the entire user datagram, (header+data)

Length: defines the total length of the user datagram, (header+data)


8/57

8

UDP length

= IP length IP headers length

Note


9/57

9

Pseudoheader for checksum calculation

The pseudoheader is a part of the header of the IP packet in which the user datagram

is to be encapsulated with some fields filled with 0s.


10/57

10

The next figure shows the checksum calculation for a

very small user datagram with only 7 bytes of data.

Because the number of bytes of data is odd, padding is

added for checksum calculation. The pseudoheader as

well as the padding will be dropped when the user

datagram is delivered to IP.

Example 1


11/57

11

Checksum calculation of a simple UDP user datagram


12/57

12

6.5: TCP6.5: TCP

TCP is a connectionTCP is a connection--oriented protocol; it createsoriented protocol; it creates

a virtual connection between two a virtual connection between two TCPsTCPs to send to send

data. In addition, TCP uses flow and error data. In addition, TCP uses flow and error

control mechanisms at the transport level.control mechanisms at the transport level.


13/57

13

Well-known ports used by TCP

Like UDP, TCP provides process to process communication using port numbers.


14/57

14

Stream delivery

TCP creates an environment in which the two processes seem to be connected by

an imaginary tube that carries their data across the Internet.

The sending process produces (write to) the stream of bytes, and the receiving

Produces consumes (reads from) them.


15/57

15

Sending and receiving buffers

TCP needs buffers for storage because the sending and receiving may not write or

read data at the same speed.


16/57

16

TCP segments

The IP layer needs to send data in packets, not as stream of bytes.

At the transport layer- TCP groups a number of bytes together into a packet:segment,

adds a header to each segment (for control purposes) and delivers the segment to the

IP layer for transmission.


17/57

17

The bytes of data being transferred ineach connection are numbered by TCP.

The numbering starts with a randomly

generated number.

Note


18/57

18

Solution:

The following shows the sequence number for each segment:

Example 2

Suppose a TCP connection is transferring a file of 5000

bytes. The first byte is numbered 10,001. What are the

sequence numbers for each segment if data are sent infive segments, each carrying 1000 bytes?


19/57

19

The value in the sequence number field of asegment defines the number of the first data byte

contained in that segment.

Note


20/57

20

The value of the acknowledgment field in asegment defines the number of the next byte a

party expects to receive.

The acknowledgment number is cumulative.

Note


21/57

21

TCP segment format


22/57

22

Control field

These bits enable flow control, connection establishment and termination,

connection abortion, and the mode of data transfer in TCP.


23/57

23

Description of flags in the control field


24/57

24

TCP Connection

Connection oriented transmission requires three phases:connection

establishment, data transfer and connection termination.

Connection establishment requires three-way handshaking;

connection termination requires three- or four-wayhandshaking with a half-close option.


25/57

25

Connection establishment using three-way handshaking

Phase 1:Phase 1:


26/57

26

A SYN segment cannot carry data, but it

consumes one sequence number.

Note


27/57

27

A SYN + ACK segment cannotcarry data, but does consume one

sequence number.

Note


28/57

28

An ACK segment, if carrying no data, consumes

no sequence number.

Note


29/57

29

Data transfer

Phase 2:Phase 2:


30/57

30

Connection termination using three-way handshaking

Phase 3:Phase 3:


31/57

31

The FIN segment consumes one sequencenumber if it does not carry data.

Note


32/57

32

The FIN + ACK segment consumesone sequence number if it

does not carry data.

Note


33/57

33

Half-close

Half-close: One end can stop sending data while still receiving data


34/57

34

Flow Control

The size of the window at one end is determined by the lesser of two values:receiver window (rwnd) or congestion window (cwnd)

The sliding window is of TCP is of variable size and a byte-oriented.


35/57

35

A sliding window is used to make transmissionmore efficient as well as to control the flow of data

so that the destination does not becomeoverwhelmed with data.

TCP sliding windows are byte-oriented.

Note


36/57

36

What is the value of the receiver window (rwnd) for host A if

the receiver, host B, has a buffer size of 5000 bytes and 1000

bytes of received and unprocessed data?

Example 3

Solution

The value of rwnd = 5000 1000 = 4000. Host B can receive only 4000 bytes of data before

overflowing its buffer. Host B advertises this value

in its next segment to A.


37/57

37

What is the size of the window for host A if the value of rwnd

is 3000 bytes and the value of cwnd is 3500 bytes?

Example 4

Solution

The size of the window is the smaller of rwnd andcwnd, which is 3000 bytes.


38/57

38

The next figure shows an unrealistic example of a

sliding window. The sender has sent bytes up to 202.

We assume that cwnd is 20 (in reality this value is thousands of bytes). The receiver has sent an

acknowledgment number of 200 with an rwnd of 9

bytes (in reality this value is thousands of bytes). The

size of the sender window is the minimum of rwnd and

cwnd, or 9 bytes. Bytes 200 to 202 are sent, but not

acknowledged. Bytes 203 to 208 can be sent without

worrying about acknowledgment. Bytes 209 and abovecannot be sent.

Example 5


39/57

39


40/57

40

Some points about TCP sliding windows:The size of the window is the lesser of rwnd and cwnd.

The source does not have to send a full window sworth ofdata.

The window can be opened or closed by the receiver, butshould not be shrunk.

The destination can send an acknowledgment at any time aslong as it does not result in a shrinking window.

The receiver can temporarily shut down the window; thesender, however, can always send a segment of 1 byte afterthe window is shut down.

Note


41/57

41

ACK segments do not consume sequence

numbers and are not acknowledged.

Note


42/57

42

In modern implementations, a retransmissionoccurs if the retransmission timer expires or three

duplicate ACK segments have arrived.

Note


43/57

43

No retransmission timer is set for an ACKsegment.

Note


44/57

44

Data may arrive out of order and be temporarilystored by the receiving TCP,

but TCP guarantees that no out-of-order segmentis delivered to the process.

Note


45/57

45

Error Control

Error control is handled by the checksum,

acknowledgement, and time-out.

Corrupted and lost segments are retransmitted, and

duplicate segments are discarded.

Data may arrive out of order and are temporarily stored by

the receiving TCP, but TCP guarantees that no out-of-

order segment is delivered to the process.

N l i


46/57

46

Normal operation


47/57

47

Lost segment


48/57

48

The receiver TCP delivers only ordered data to theprocess.

Note


49/57

49

Fast retransmission


50/57

50

Congestion control refers to techniques and mechanisms

that can either prevent congestion, before it happens, orremove congestion, after it has happened.

Congestion Control


51/57

51

Slow start, exponential increase


52/57

52

In the slow-start algorithm, the size of thecongestion window increases exponentially until

it reaches a threshold.

Note


53/57

53

Congestion avoidance, additive increase


54/57

54

In the congestion avoidance algorithm, the size ofthe congestion window increases additively until

congestion is detected.

Note


55/57

55

An implementation reacts to congestion

detection in one of the following ways:If detection is by time-out, a new slow start

phase starts.If detection is by three ACKs, a new

congestion avoidance phase starts.

Note


56/57

56

TCP congestion policy summary

TCP vs UDP


57/57

57

TCP vs. UDP

TCP provides a reliable connection-oriented service,

whereas UDP provides an unreliable connectionless

service by delivering the UDP datagrams on a best-effort basis.

UDP is simpler than TCP. There is no connection, flow

control, congestion control, sequencing, packet

acknowledgement, and reliability required by UDP

transmission.

datacommchapter 6 part 2

Documents