Document

CS 40024: Internet Technology
Full Marks: 90
Time: 3 Hours
End-Semester Examination
Date: 22-04-2015
Note:
(i) There are total SEVEN questions in this paper. Answer any SIX questions. The answers
should be precise and to-the-point.
(ii) Write down the assumptions clearly, if any. No queries will be entertained during the exam
hours.
1. (a) An ISP is granted a block of addresses starting with 190.100.0.0/16. The ISP needs to
distribute these addresses to three groups of customers as follows:
• The first group has 60 customers; each need 250 addresses.
• The second group has 124 customers; each need 120 addresses.
• The third group has 124 customers, each need 60 addresses.
Design the sub-blocks with clearly specifying the IP ranges for every customer in each
groups. How many addresses are still available after these allocations?
[10]
(b) An ISP has a block of 1024 addresses. It needs to divide the addresses to 1024 customers, does it need subnetting? Explain your answer. [Answer within 50 words]
[3]
(c) An IP datagram arrives with fragmentation offset 0 and more fragment bit 0. Is this a
first fragment, a middle fragment or the last fragment? Explain your answer. [Answer
within 50 words]
[2]
2. (a) An IP datagram is carrying a TCP segment destined for address 130.14.16.17/16. The
destination address gets corrupted and it arrives at destination 130.14.16.19/16. How
does the receiving TCP react to this error? [Answer within 50 words]
[3]
(b) The sequence number field in the TCP header is 32 bits long, which is big enough to
cover 4 billion bytes of data. Even if this many bytes are never transferred over a single
connection, why might the sequence number still wrap around 232 − 1 to 0? [Answer
within 50 words]
[3]
—– Page 1 of 4 —–
(c) Let you have to design a reliable byte-stream protocol that uses a sliding window. This
protocol will run over a 1 Gbps network. The RTT of the network is 140 ms, and the
maximum segment lifetime is 60 seconds. How many bits would you include in the
advertised window and sequence number fields of your protocol header?
[5]
(d) Assume that a mobile node moves from a 1 Mbps bandwidth network to a 100 Mbps
bandwidth network. Explain how this migration can affect TCP congestion control
behavior? What will happen if the mobile node moves from a 100 Mbps bandwidth
network to a 1 Mbps bandwidth network? [Answer within 60 words]
[4]
3. (a) Explain why congestion avoidance is necessary for providing QoS when TCP can control congestion? Give two specific reasons. [Answer within 50 words]
[4]
(b) You need to design an application for interactive audio streaming with on demand file
transfer support over a 10 Mbps Ethernet network. Specify what protocols you’ll use
at the following layers of the network protocol stack - data link, network and transport.
Specify the protocols at the end hosts and at the intermediate routers. Give specific
reasons behind the selection of each protocol. Do you need to customize any protocol?
If yes, then explicitly specify what customizations are required. You can use following
table structure to answer this question.
[8]
Layer
Data Link
Data Link
Data Link
Network
Transport
Location
End Host
Router
...
...
...
Protocols Reasons
Protocol 1
...
Protocol 2
...
...
...
...
...
...
...
Customizations
...
...
...
...
...
(c) Explain why traffic shaping is required for supporting QoS? Explain a method for implementing traffic shaper for differentiated service architecture. [Answer within 50
words]
[3]
4. (a) Show that the bandwidth share ri , for a specific flow i passing through a weighted fair
queuing (WFQ) router is given by,
wi
ri = P
R
n
wj
j=1
where R is the total bandwidth, and wi is the weight of ith flow. There are a total of n
number of flows passing through the router.
[5]
(b) Assume that this flow has burst size b. Derive an expression for the maximum delay
experienced by this flow.
[5]
—– Page 2 of 4 —–
(c) Assume that there are two subnets SA and SB connected via an edge router R. Packets
are flowing from subnet SA to subnet SB . To avoid congestion, you want to implement
random early detection (RED) algorithm. Where should you implement the RED algorithm to have minimal changes in the routers - (a) Routers in subnet SA , (b) Routers in
subnet SB , (c) Edge router R, (d) Routers in subnet SA and edge router R, (e) Routers
in subnet SB and edge router R? Justify your answer. [Answer within 50 words]
[5]
5. (a) Explain the differences between expedited forwarding (EF) and assured forwarding
(AF) per hop behaviors. Give one example when you should use EF and another example when you should use AF. [Answer within 100 words]
[6]
(b) What is the function of bandwidth broker? Do we need a bandwidth broker in integrated
service architecture? Why or why not? [Answer within 100 words]
[3]
(c) With the help of a diagram, explain how level distribution protocol (LDP) works in
MPLS? Can you implement a MPLS tunnel over an IP network? Explain with an example. [Answer within 200 words]
[6]
6. (a) Suppose a router has three input flows and one output. It receives the packets given
in the following table, all at about the same time, in the order listed, during a period
in which the output port is busy but all queues are otherwise empty. Give the order in
which the packets are transmitted assuming:
(i) Fair queuing
(ii) Weighted fair queuing, with flow 2 having twice as much share as flow 1, and flow
3 having 1.5 times as much share as flow 1. Note that ties are to be resolved in
order flow 1, flow 2, flow 3.
Packet
1
2
3
4
5
6
7
8
Size
200
200
160
120
160
210
150
90
Flow
1
1
2
2
2
3
3
3
[10]
(b) In HTTP version 1.0, a server marked the end of a transfer by closing the connection.
Explain why, in terms of the TCP layer, this was a problem for the servers. How can
you solve this problem? [Answer within 100 words]
[3]
(c) Explain, why mail server uses two different protocols - one for transferring and another
for retrieving emails? [Answer within 30 words]
[2]
—– Page 3 of 4 —–
7. (a) It is said that IPSec may not work with network address translation (NAT). However,
whether IPSec will work with NAT depends on which mode of IPSec and NAT we
use. Suppose we use true NAT, where only IP addresses are translated (without port
translation). Will IPSec and NAT work in each of the following cases? Explain why
or why not.
(i) IPSec uses ESP transport mode.
(ii) IPSec uses ESP tunnel mode.
[Answer within 200 words]
[7]
(b) Consider the following simple UDP protocol for downloading files;
•
•
•
•
Client sends a file requests
Server replies with first data packet
Client sends ACK
The server and the client proceed using stop-and-wait.
Suppose client and server possess keys KC and KS , respectively. These keys are known
to each other.
(i) Extend the file downloading protocol, using the keys and message authentication
code (MAC), to provide sender authentication and message integrity. Your protocol should also be resistant to replay attacks.
(ii) How does extra information in your revised protocol protect against arrival of late
packets from prior connection incarnations and sequence number wraparound?
[Answer within 300 words]
[8]
—– Page 4 of 4 —–