Solution and Evaluation Scheme (CN) 2019 Autumn

1. (a) ARP
(b) No, the receiver cannot be absolutely certain that no bit errors have occurred. This is because of
the manner in which the checksum for the packet is calculated. If the corresponding bits (that would
be added together) of two 16-bit words in the packet were 0 and 1 then even if these get flipped to 1
and 0 respectively, the sum still remains the same. Hence, the 1s complement the receiver calculates
will also be the same. This means the checksum will verify even if there was transmission error.
(c) Looking into the Source IP address of the received packets.
(d) 4 bytes as min=46 bytes
(e) Caching reduces bandwidth consumption; therefore, it decreases network traffic and diminishes
network congestion. Caching reduces the workload of the remote web server.
(f) Yes, Depends on the no. of interfaces/ports available in the router
(g) The 8-bit protocol field in the IP datagram contains information about which transport layer
protocol the destination host should pass the segment to.
(h) HLEN keeps the length of TCP header as it varies from 20-60 bytes. If it does not include any
options and padding then it's value is 5.
(I) When a new connection starts, the sequence number is set to a random value, to help avoid
interference between successive incarnations of the same connection. As a result, the sequence
number can wrap around after only a few packets, if you happen to start with a number close to ffff
ffff.
(j) Yes. Delayed Acknowledgements fall outside the current window.

2. (a) Step Marking should be done based on the steps carried out by the students.
Step-01: Frames through 0 to 4 have been transmitted - 4 , 3 , 2 , 1 , 0
Step-02: 0 times out. So, sender retransmits it - 0 , 4 , 3 , 2 , 1
Step-03: 5 (a new frame) is transmitted - 5 , 0 , 4 , 3 , 2 , 1
Step-04: 1 times out. So, sender retransmits it - 1 , 5 , 0 , 4 , 3 , 2
Step-05: 2 times out. So, sender retransmits it - 2 , 1 , 5 , 0 , 4 , 3
Step-06: 6 (another new frame) is transmitted - 6 , 2 , 1 , 5 , 0 , 4 , 3
Finally, the outstanding packets in sender’s window are : 6 , 2 , 1 , 5 , 0 , 4 , 3
(b) Explanation of Count to infinity problem – 1 mark, Solution to the problem – 1 mark
Comparison of OSPF and RIP – 2 marks

Q3 a) Fields in IP header for fragmentation – [2Marks]

• Identification (16 bits) – use to identify fragments of same frame.
• Fragment offset (13 bits) – use to identify sequence of fragments in the frame. It generally
indicates number of data bytes preceeding or ahead of the fragment.
Maximum fragment offset possible = (65535 – 20) – 1 = 65514
{where 65535 is maximum size of datagram and 20 is minimum size of IP header}
So, we need ceil(log265514) = 16 bits for fragment offset but fragment offset field has only
 13 bits. So, to represent efficiently we need to scale down fragment offset field by 216/213 =
8 which acts as a scaling factor. Hence, all fragments except the last fragment should have
data in multiples of 8 so that fragment offset ∈ N.
• More fragments (MF = 1 bit) – tells if more fragments ahead of this fragment i.e. if MF = 1,
more fragments are ahead of this fragment and if MF = 0, it is the last fragment.
• Don’t fragment (DF = 1 bit) – if we don’t want the packet to be fragmented then DF is set
i.e. DF = 1.
Reassembly of Fragments – [2Marks]
It takes place only at destination and not at routers since packets take independent path(datagram
packet switching), so all may not meet at a router and hence a need of fragmentation may arise
again. The fragments may arrive out of order also.
Q3 b) It takes 1 RTT to increase CongWin to 6 MSS; 2 RTTs to increase to 7 MSS; 3 RTTs to
increase to 8 MSS; 4 RTTs to increase to 9 MSS; 5 RTTs to increase to 10 MSS; and 6 RTTs to
increase to 11 MSS. [4Marks]

4.(a) Explanation – 3 marks

• In TCP client/server communication either end can enter into the TIME_WAIT state during
connection tear down process.
• Which ever end initiates the connection tear down process that end enters into the
TIME_WAIT state.
• Assume TCP_Client initiates the close thereby going into the TIME_WAIT state and has to
wait for 2MSL time period before going to CLOSED state and making the port available for
reuse. But as the TCP_client uses a ephemeral port this is not going to pose any problem
during the restart of the TCP_Client.
• Assume TCP_Server initiates the close thereby going into the TIME_WAIT state and has to
wait for 2MSL time period before going to CLOSED state and making the port available for
reuse. But the TCP_server uses a well known port. Hence, during the restart of the
TCP_Server before 2MSL time period of the TIME_WAIT state it is going to say bind
failure. Hence as a consequence of the same you can not start the server in the same port i.e.
either one has to wait for 2 MSL time period to restart the TCP server or change the server
port for the same.
Situation When this will occur (1 Mark)
• If some thing goes wrong during the communication between the TCP client and server and
as a result we have to forcefully stop the TCP client and server. While doing so the user
forcefully stops the TCP server ahead of TCP client then this situation might occur.

(b) Functionality of each layer – 1 Mark., Protocols used in each layer – 1 Mark.
Services provided by each layer – 1 Mark, How it is different from OSI model – 1 Mark.
5. (a) Broadcast vs Multicast – 2 Marks, How Limited broadcast is different from Directed
broadcast – 1 Mark
Limited Broadcast – 255.255.255,255
Directed Broadcast – Broadcast id of the respective network Ex. If the netwwork id is
192.168.1.0/24 then the broadcast id will be 192.168.1.255
Scenario where Limited broadcast will not help – 1 Mark
Let us consider a scenario where an user wants to broadcast a message to those who are not present
in the same network as the sender is here limited broadcast will not help rather directed broadcast
will be used to carry out the task.
(b) Explanation of Ethernet Frame format – 2 Marks
CSMA/CD explanation – 2 Marks

6. (a) Step Marking should be given based on the steps carried out by the student.
Since the link capacity is only 100 Mbps, so host A’s sending rate can be at most 100Mbps.
• Still, host A sends data into the receive buffer faster than Host B can remove data from the
buffer.
• The receive buffer fills up at a rate of roughly 40Mbps.
• When the buffer is full, Host B signals to Host A to stop sending data by setting
RcvWindow = 0.
• Host A then stops sending until it receives a TCP segment with RcvWindow > 0.
• Host A will thus repeatedly stop and start sending as a function of the RcvWindow values it
receives from Host B.
• On average, the long-term rate at which Host A sends data to Host B as part of this
connection is no more than 60Mbps.
(b) Description of the various components of routers – 2 Marks, Scenario where packet loss
can occur at the input ports of a router – 1 Mark, Solution to packet loss – 1 Mark
Various components of router
i. Input port
ii. Output port (Explain in brief, Refer text book)
iii. Switching fabric
iv. Routing processor
- Scenarios where packet loss can occur at input port is the queue that may form at the input
port.
- Queuing mainly depends on factors like
a. Traffic load
b. Relative speed of the switching fabric
c. Line speed
- If queue grow large, router’s memory will be eventually exhausted and packet loss will
occur.
- Packet loss at the i/p port can be eliminated by increasing the relative speed of switching
fabric.
7. (a) Why does an application using UDP have more control of what data is sent in a
segment? 2 Marks
UDP is message oriented while TCP is byte oriented. If we send a message in TCP there is no
control from the user side that the message will go as one message. It may be sent as a part of the
whole communication or it may be sent by segmenting the message in to several messages. But in
case of UDP if the user sents a message it will be sent and receibed as a independent message.
Why does an application using UDP have more control on when the segment is sent? 2 Marks
In case of TCP when a user sent a message there is no guarantee that the message will be sent
immediately if the Nagel's algorithm is on i.e. the message will be sent as per the rules of Nagel’s
algorithm to control the flow and congestion in the network. But as in UDP there is no flow and
congestion control hence when a user sent a message it will be sent immediately to the receiver end.
(b) Step Marking should be given based on the steps carried out by the student.
The window size (in bytes) must be RTTx Bandwidth = 1OA9/8 x 0.14 = 17500000 bytes. So,
we need, 25 bits for the advertised window (allows a max window size of 33554431 bytes). In 60
sees, 19A9/8 x 60 =7500000000 bytes can be transmitted. They must all have unique sequence
numbers. So, we need 33 bits for the sequence numbers.
8.(a) Functionality of DHCP Server - 2 marks
• IP Address to an interface dynamically
• subnet mask of the assigned IP Addressing
• Default Gateway IP Address
• Name server/ Local DNS Server IP Address
Need of Running DHCP client on a well know port – 2 Marks
• Let's assume that host A is using the DHCP client on ephemeral port 5000, and host B
(which is on the same network) is using SMTP client on the same port.
• Now when the DHCP server sends a broadcast reply message with the broadcast IP address
255.255.255.255 and destination port no. 5000, then host A will accept the correct message
on its DHCP client on the application layer. But the SMTP client which is running on the
application layer of host B will get an incorrect message.
• The use of an well-known port (in our case 68) prevents the use of the same two destination
port numbers and hence it prohibits other protocols from using the same port which is
already in use by another protocol.
• In simple words, it prevents an application from getting a message from a completely
different protocol.
(b) For a given Class A network represented as 10.0.0.0, the requirement of host per each
subnetwork is 500. Find the following information:
(i) Customized subnet Mask used : 255.255.254.0 [Masking 15 bit host ]
(ii) Number of Hosts in each subnet : 510 host/subnet (29-2)
(iii) Range of valid IP numbers that can be assigned in each subnetwork:
IP ranges: 10. 0. 0. 0 to 10. 0. 1. 255, [First subnetwork IP]
10. 0. 2. 0 to 10. 0. 3. 255 [Second subnetwork IP]
10. 0. 4. 0 to 10. 0. 5. 255 [Third subnetwork IP]

…………………………
…………………………
Fully correct 4 marks, partially correct, 1 mark for each bit.