1. Transport Layer Our goals:
understand principles behind transport layer services:
multiplexing/demultiplexing
reliable data transfer
flow control
congestion control
Services of transport layer are constrained by N/W layer.
learn about transport layer protocols in the Internet:
UDP: connectionless transport
TCP: connection-oriented transport
TCP congestion control
TCP : Segment
UDP: Datagram

2. Transport services and protocols
provide logical communication between app processes running on different hosts
transport protocols run in end systems
sending side: breaks app messages into segments, passes to network layer
receiving 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
network
data link
physical
network
data link
physical
network
data link
physical
logical end-end transport
network
data link
physical
network
data link
physical
application
transport
network
data link
physical

3. Transport vs. network layer
network layer: logical communication between hosts
transport layer: logical communication between processes
relies on, enhances, network layer services
IP (Internet Protocol) is a Network Layer protocol
Unreliable
Best Effort Service
No ordering of segments
No guaranteed delivery

4. Internet transport-layer protocols
reliable, in-order delivery (TCP)
congestion control
flow control
connection setup
Integrity checking
unreliable, unordered delivery: UDP
no-frills extension of “best-effort” IP
services not available:
delay guarantees
bandwidth guarantees
application
transport
network
data link
physical
network
data link
physical
network
data link
physical
network
data link
physical
logical end-end transport
network
data link
physical
network
data link
physical
application
transport
network
data link
physical

5. Multiplexing/demultiplexing
Multiplexing at send host:
Demultiplexing at rcv host:
delivering received segments
to correct socket
gathering data from multiple
sockets, enveloping data with
header (later used for
demultiplexing)
= socket
= process
application
transport
network
link
physical P1 P2 P3 P4
host 1
host 2
host 3

6. Connectionless demultiplexing
Create sockets with port numbers:
DatagramSocket mySocket1 = new DatagramSocket(49911);
DatagramSocket mySocket2 = new DatagramSocket(49922);
UDP socket identified by two-tuple:
(dest IP address, dest port number)
When host receives UDP segment:
checks destination port number in segment
directs UDP segment to socket with that port number
IP datagrams with different source IP addresses and/or source port numbers directed to same socket

7. Connectionless demux (cont)
DatagramSocket serverSocket = new DatagramSocket(6428); P2 P1 P1 P3
SP: 6428
DP: 9157
SP: 6428
DP: 5775
SP: 9157
DP: 6428
SP: 5775
DP: 6428
client
IP: A
Client
IP:B
server
IP: C
SP provides “return address”

8. Connection-oriented demux
TCP socket identified by 4-tuple:
source IP address
source port number
dest IP address
dest port number
recv host 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

9. Connection-oriented demux (cont)P1 P4 P5 P6 P2 P1 P3
SP: 9157
DP: 80
S-IP: A
D-IP:C
SP: 5775
DP: 80
D-IP:C
S-IP: B
SP: 9157
DP: 80
D-IP:C
client
IP: A
Client
IP:B
server
IP: C
S-IP: B

10. Connection-oriented demux: Threaded Web ServerP1 P4 P5 P6 P2 P1 P3
SP: 5775
DP: 80
S-IP: B
D-IP:C
SP: 9157
SP: 9157
DP: 80
client
IP: A
DP: 80
Client
IP:B
server
IP: C
S-IP: A
S-IP: B
D-IP:C
D-IP:C

11. UDP

12. Position of UDP in the TCP/IP protocol suite

13. UDP: User Datagram Protocol [RFC 768]
“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

14. User datagram format Application data (message) UDP segment format
source port #
dest port #
32 bits
Application
data
(message)
UDP segment format
length
checksum
Length, in
bytes of UDP
segment,
including
header

15. UDP Cont… often used for streaming multimedia apps loss tolerant
rate sensitive
reliable transfer over UDP: add reliability at application layer
application-specific error recovery!

16. UDP Advantages No extra processing of data.
No connection establishment (which can add delay)
Simple: no connection state at sender, receiver
Small packet header overhead
No congestion control: UDP can blast away as fast as desired

17. The following is a dump of a UDP header in hexadecimal format.
a. What is the source port number?
b. What is the destination port number?
c. What is the total length of the user datagram?
d. What is the length of the data?
e. Is the packet directed from a client to a server or vice versa?
f. What is the client process?

18. Solution
a. The source port number is the first four hexadecimal digits (CB84)16 or
b. The destination port number is the second four hexadecimal digits (000D)16 or 13.
c. The third four hexadecimal digits (001C)16 define the length of the whole UDP packet as 28 bytes.
d. The length of the data is the length of the whole packet minus the length of the header, or 28 – 8 = 20 bytes.
e. Since the destination port number is 13 (well-known port), the packet is from the client to the server.
f. The client process is the Daytime

19. Given UDP header in hexadecimal format: 0632 000D 001C E217
i) Source port number? ii) Destination port number? iii) Total length of user datagram? iv) Length of data? What is client process? v) Is the packet directed from a client to server or vice-versa?
Given UDP  header in hexadecimal format: D 00 1C E2 17
1. The source port number is that is 1586 in decimal
2. The destination port number is that is 13
3. The total length of the user datagram is that is 28bytes
4. The length of the data = 28-8 = 20 bytes. The client process is Daytime because
destination port number(Destination port number represents Daytime process)
5. The packet is directed from a client to a server because destination port number
is 13.

20. UDP checksum
Goal: detect “errors” (e.g., flipped bits) in transmitted segment
Sender:
treat segment contents as sequence of 16-bit integers
checksum: addition (1’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 ….

21. Internet Checksum Example
Note
When adding numbers, a carryout from the most significant bit needs to be added to the result
Example: add two 16-bit integers
Kurose and Ross forgot to say anything about wrapping the carry and adding it to low order bit
+ 1
wraparound
sum
checksum

23. s

24. Sliding Window
The seq number are modulo 2m, hence seq no= (0 to 2m -1).
Say m=4 and size of window is 7.

25. Flow and Error Control Mechanisms
Stop-and-Wait Protocol
Go-back-N Protocol
Selective Repeat Protocol

26. Stop-and Wait Protocol
Sender and receiver use window size of 1
The sender sends one packet at a time and waits for an ack before sending the next one
If pkt lost or corrupted receiver silently discards pkt.
After time-out sender retransmits packet

28. Stop-and-Wait ARQ, damaged frame
When a receiver receives a damaged frame, it discards it and keeps its value of R.
After the timer at the sender expires, another copy of frame 1 is sent.

29. Stop-and-Wait, lost ACK frame
If the sender doesn’t receives an ACK (or damaged ACK, it discards it)
When the timer of the sender expires, the sender retransmits frame 1.
Receiver has already received frame 1 and expecting to receive frame 0 (R=0). Therefore it discards the second copy of frame 1.

30. Stop-and-Wait, delayed ACK frame
The ACK can be delayed at the sender due to some problem
It is received after the timer for frame 0 has expired.
Sender retransmitted a copy of frame 0. However, R =1 means receiver expects to see frame 1. Receiver discards the duplicate frame 0.
Sender receives 2 ACKs, it discards the second ACK.

32. Piggybacking A method to combine a data frame with ACK.
Station A and B both have data to send.
Instead of sending separately, station A sends a data frame that includes an ACK.
Station B does the same thing.
Piggybacking saves bandwidth.

33. Stop-and-Wait implements:
Flow Control
by forcing the sender to wait for an ack
Error Control
by discarding corrupted pkts and letting the sender resend unacked packets on timer expires.
Inefficient
Low channel utilization

34. Go-back-N Protocol Sequence number Ack number
0 to 2m-1, where m is bits for seq. number
Ack number
is cumulative and defines the sequence number of the next packet expected to arrive

35. Send Window The maximum size of the window is 2m − 1
Say, sliding window of size 7 (m = 3) for the Go-Back-N protocol.

37. Receive Window size of the receive window is always 1
When a correct packet is received, the window slides, Rn = (Rn + 1) modulo 2m
Timer expires
Ack received
Data received from app
and sent for 1st time

39. Timer not set for
discarded Acks

41. Selective Repeat ARQ, sender and receiver windows
Go-Back-N ARQ simplifies the process at the receiver site. Receiver only keeps track of only one variable, and there is no need to buffer out-of-order frames, they are simply discarded.
However, Go-Back-N ARQ protocol is inefficient for noisy link. It is bandwidth inefficient and slows down the transmission.
In Selective Repeat ARQ, only the damaged frame is resent. More bandwidth efficient but more complex processing at receiver.
It defines a negative ACK (NAK) to report the sequence number of a damaged frame before the timer expires.
Maximum size of the send window is: 2m-1.
For example, if m = 4, the sequence numbers go from 0 to 15, but the maximum size of the window is just 8 (it is 15 in the Go-Back-N Protocol).

42. Outstanding packet: waiting for ACK
Timer: separate timer for each outstanding pkt.
Say m=2

44. Assume a sender sends 6 packets: packets 0, 1, 2, 3, 4, and 5
Assume a sender sends 6 packets: packets 0, 1, 2, 3, 4, and 5. The sender receives an ACK with ackNo = 3. What is the interpretation if the system is using GBN or SR?

45. If the system is using GBN, it means that packets 0, 1, and 2 have been received uncorrupted and the receiver is expecting packet 3. If the system is using SR, it means that packet 3 has been received uncorrupted; the ACK does not say anything about other packets.

47. If we receive a corrupted frame and a NAK has not yet been sent, we send a NAK to tell the other site that we have not received the frame we expected.
The first NAK sent is remembered (using the nakSent variable) and is not sent again until the frame slides.
A NAK is sent once for each window position and defines the first slot in the window.

48. The next point is about the ACKs.
In Selective Repeat, ACKs are sent when data are delivered to the upper layer.
If the data belonging to n frames are delivered in one shot, only one ACK is sent for all of them.
Individual timer is set for each segment.