eusp

THE UNIVERSITY OF THE

SOUTH PACIFIC

C5310: Computer Networks

School of Computing, Information and Mathematical Sciences

Final Examination
Semester 1 2017

Mode: Face to Face

Duration of Exam: 3 hours + 10 minutes

Reading Time: 10 minutes

Writing Time: 3 hours

Instructions:

1. This is a closed book exam

2. This exam has four sections:
Section I: Multiple Choice Question (18 Marks)
Section II: Matching Question (14 Marks)
Section III: Short Answer (48 Marks)
Section IV: Problem Solving (20 Marks)
3. All questions are compulsory.
4. Write your answers in this booklet in the space provided.
5. The full marks of this exam are 100 marks, which are worth 50% of your overall mark.
6. Number of pages (including this cover page): 23

Name: ___________________________________

ID:
Section I: Multiple Choice Question [18 Marks]

Circle the correct answer in the grid provided below for Questions 1-18.

Question Answer
1 a b c d e
2 a b c d e
3 a b c d e
4 a b c d e
5 a b c d e
6 a b c d e
7 a b c d e
8 a b c d e
9 a b c d e
10 a b c d e
11 a b c d e
12 a b c d e
13 a b c d e
14 a b c d e
15 a b c d e
16 a b c d e
17 a b c d e
18 a b c d e

1. Unnecessary retransmission of redundant packets will be caused by _ __

a. packet loss
b. packet error
c. packet overflow
d. premature timeout
e. none of the above

2. Which of the followings is not applicable for TCP?

a. full duplex
b. fast retransmit
c. best effort
d. point-to-point
e. none of the above

3. Which of the followings is not applicable for UDP?

a. identified by four-tuple
b. best effort
c. connectionless
d. support error detection with checksum
e. none of the above

Page 2 of 23
4. A reliable transfer protocol can detect if a received packet is a duplicate by using
a. timeout
b. sequence number
c. checksum
d. TTL
e. none of the above

5. When switching via _ _-" multiple packets can be switched simultaneously.

a. memory
b. crossbar
c. bus
d. token
e. none of the above

6. A NAT device will not replace _ _ for incoming datagrams

a. destination IP address
b. source port number
c. destination port number
d. checksum
e. none of the above

7. The technique that allows a local network to use just one IP address to connect to the outside world is
called
a. DHCP
b. DNS
c. NAT
d. AIMD
e. none of the above

8. Utilizing ___ bandwidth can allow P2P model to perform better than client/server model for file
distribution.
a. server download
b. server upload
c. peer download
d. peer upload
e. none ofthe above

9. Which of the followings can be used to measure the rate (bits/time unit) at which bits transferred
between sender and receiver?
a. loss rate
b. latency
c. bandwidth
d. traffic intensity
e. none of the above

Page 3 of 23
10. Which of the followings is the type of the answer to a DNS query for the IP address of a domain
name?
a. Type A
b. Type NS
c. Type (NAME
d. Type MX
e. none of the above

11. Which of the followings is applicable for IPv4 and IPv6?

a. IPv6 upgrades IPv4. Therefore, IPv6 can be used in all network devices.
b. Network cannot operate with mixed IPv4 and IPv6 routers.
c. IPv4 datagram can be carried as payload in IPv6 datagram among IPv6 routers.
d. IPv6 datagram can be carried as payload in IPv4 datagram among IPv4 routers.
e. none of the above

12. In hierarchical routing networks,

a. routers in the same AS can run different intra-AS routing protocol
b. routers in different AS can run the same intra-AS routing protocol
c. different AS must run the same inter-AS routing protocol
d. different AS must run different inter-AS routing protocol
e. none of the above

13. In hierarchical routing networks, a router's forwarding table is configured by

a. the intra-AS routing algorithm used by the router's AS
b. the inter-AS routing algorithm used by the router's AS
c. the intra-AS and the inter-AS algorithms used by the router's AS
d. the routing algorithm used by the gateway routers
e. none of the above

14. Reverse path forwarding is a mechanism used to

a. control the flooding of broadcast packets
b. forward packets back to the sender using the path the received packet travelled.
c. filter out packets that did not arrive on the shortest path between the source and the receiver
d. remove duplicate packets
e. none of the above

15. Link layer is implemented in

a. the process of hosts, routers and switchers
b. network interface card attached into host's system buses
c. router's switch fabric
d. switcher's core component
e. none of the above

Page 4 of 23
16. Ethernet is
a. unreliable as the receiving NIC does not transmit any ACKs to the sending NIC
b. reliable as data in dropped frames can be detected and recovered by using higher layer
protocols
c. unreliable as it does not use TCP and thus no handshaking between the sending NIC and the
receiving NIC
d. reliable as it is a dominant wired LAN technology
e. none of the above

17. In wireless networks,

a. reliable data transfer cannot be achieved as wireless networks are not secured
b. reliable data transfer cannot be achieved as wireless networks do not implement the transport
layer, therefore, rdt protocols cannot be used.
c. reliable data transfer cannot be achieved as terminals can be hidden from each other in terms of
signal attenuation.
d. reliable data transfer is difficult to achieve as multiple senders and receivers can create
additional problems beyond multiple access.
e. none of the above.

18. When the efficiency of CSMA/CD is defined as lithe long-run fraction of time during which frames are
being transmitted on the channel without collision when there is a large number of active nodes,
with each node having a large number of frames to send", the approximation of efficiency can be
stated by using the following formula:

1
efficiency - - - - - - - -
1 + 5t prop/ttrans
where
tprop = max prop delay between 2 nodes in LAN
ttrans = time to transmit max-size frame

a. According to the formula, efficiency goes to 1 as t prop goes to O. However, this cannot reflect the
efficiency since when propagation delay is 0, colliding nodes will abort immediately and
therefore the channel is wasted.
b. According to the formula, efficiency goes to 1 as ttrans goes to infinity. However, this cannot
reflect the efficiency since when a frame grabs the channel, it will hold on to the channel for a
very long time and therefore the channel will not be doing productive work.
c. Efficiency calculated from the formula can reflect the following case: the longer the time for a
signal to propagate is, the less chance a carrier-sending node is able to sense a transmission
already begun at another node.
d. all of the above
e. none of the above

Page 5 of 23
Section II: Matching Question [14(2+3+2+3+4) Marks]

Chose the correct option provided below to fill in the blanks in the following 5 questions.

a. a TCP connection

b. an eBGP session

c. an iBGP session

d. a prefix

e. an AS-PATH

f. a NEXT-HOP

g. all BGP messages sent over the TCP connection

h. all routers in the same AS

i. all other routers in the same AS

j. all gateway routers in the neighbouring AS

k. start_timer

I. stop_timer

m. timeout

n. sndpkt

o. done

p. sent ACKed

q. sent not yet ACKed

r. not usable

s. protocol number

t. port number

u. incoming segment to sender

v. outgoing segment from sender

w. none of the above

Page 6 of 23
1. A BGP session includes _ _ _ _ and _ __

2. When a gateway router receives prefixes from _ _- " it uses _ _ to distribute the prefixes
to _ _

3. While _ _ binds the Network Layer and the Transport Layer together, _ _ binds the
Transport Layer and the Application Layer together.

4. For rdt 3.0 sender showed in Figure 1 below,

ACTION_l that will be performed is _ __
EVENT_l that will be triggered is _ _ ;
TO_BE_SENT is _ _

rdt_send(data}
rdt_rcv(rcvpkt} &&
\ =
sndpkt makeJ)kt(O, data, checksum} ( corrupt(rcvpkt) II
\ udCsend(sndpkt} isACK(rcvpkt,1} }
rdt rcv(rcvpkt} \ ACfiOJr_l A
A
timeout
udCsend(sndpkt}
start_timer
rdCrcv(rcvpkt}
&& notcorrupt(rcvpkt} rdCrcv(rcvpkt}
&& isACK(rcvpkt,1} && notcorrupt(rcvpkt}
stop_timer && isACK(rcvpkt,O}

rdt_rcv(rcvpkt}
rdt_send(data} A
=
sndpkt make...,pkt(1, data, checksum}
udt_send('.rO_BB_SBR'.E'}
start_timer

Figure 1

Page 7 of 23
5. For rcp sender's sequence number space showed in Figure 2 below,
A indicates
B indicates _ __
C indicates _ __
D indicates _ __

vvindovv
,- ~ -+----~~

I1I111111
sender sequence number space
111111111
----~----- I
B usable not
but not usable
yet sent

Figure 2

Page 8 of 23
Section III: Short Answer [48(6+4+2+5+4+8+6+2+5+6) Marks]

1. What is queuing delay? How can it be improved? [6 marks]

2. Compare and contrast RIP and OSPF. [4 marks]

Page 9 of 23
3. Do you agree that BGP is the glue that holds the Internet together? Justify your answer. [2 marks]

4. What is the effect of out-of-order delivery on Go-Back-N, Selective Repeat and TCP? [5 marks]

Page 10 of 23
5. Indicate the next hop in the following scenarios. [4(1+2+1) marks]

a. Suppose router 1c received a route as follows:

Prefix: 201.44.13/22; AS-PATH: AS3 AS131; NEXT-HOP: 138.16.64.1
Clearly indicate the location of the next hop in Figure 3 below.

Figure 3

b. Host 10.0.0.1 sends a request datagram to host 128.119.40.186, 80. The response datagram from
Host 128.119.40.186 contains a destination address 138.76.29.7, 5001. The datagram arrives at
host 10.0.0.1 has source address 128.119.40.186,80 and destination address 10.0.0.1, 3345.

Clearly indicate the location of 138.76.29.7 in Figure 4 below.

rest of - - - - _ +---- local network

Internet (e.g., home network)
10.0.0/24 10.0.0.1

10.0.0.2

10.0.0.3

Figure 4

Page 11 of 23
c. Clearly indicate the position of 24.128.190.197 in Figure 5 below using the Wireshark
captured data provided in Figure 6.

Figure 5

Figure 6

Page 12 of 23
6. Consider the TCP congestion control in Figure 7 below. [8(6+2) marks]

Congestion Windo_
(CWND)

1 11 1 1 1 Time
A Be D E F G

Figure 7

a. Each of the marked points (i.e., A, B, C, D, E, F, G) along the timeline in Figure 7 (above) indicates
relevant phase and corresponding events and actions of TCP congestion control. These phases,
events and actions can be found in the FSM Description of TCP Congestion Control presented in
Figure 8 (below). For each of the marked points in the Figure 7 (above), circle or tick the
appropriate phase, events and actions, and clearly indicate the corresponding marked points in
Figure 8 (below).

newACK
duplicate ACK

SJ
cwnd =cwnd + MSS .(MSSlcwnd)
dupACKcount++ new ACK dupACKcount = 0

A
O cwnd = cwnd+MSS
dupACKcount = 0
tmnsmit new segment(s), as ai/owed

;,..("\tmnsmit new segment(s), as allowed ....•.....'

cwnd = 1 MSS
. ssthresh = 64 KB .\.) cwnd ~ ssthresh ' ; .
: _t1~~~!'!)~:: g... A
..
timeout
ssthresh =cwndI2
cwnd= 1 MSS
dupACKcount = 0
ssthresh =cwndI2 retmnsmit missing segment
cwnd= 1 MSS
dupACKcount = 0
retmnsmit missing segment
timeout
ssthresh = cwndI2
cwnd = 1 NewACK
dupACKcount = 0
retmnsmit missing segment cwnd = ssthresh dupACKcount == 3
dupACKcount == 3 dupACKcount =0
ssthresh= cwndI2 ssthresh= cwndI2
cwnd = ssthresh + 3 cwnd = ssthresh + 3
retmnsmit missing segment retmnsmit missing segment

cwnd = cwnd + MSS

tmnsmit new segment(s), as a/lowed

Figure 8

Page 13 of 23
 b. Assume CWND is 10,000 right before F. What is the value of ssthresh at G?

7. Suppose a laptop is connected to Ethernet and sends a DNS query message. What is the sequence of
packet header on this request packet (that contains the DNS query) as it leaves the laptop starting
from the outer most header? What are their corresponding layers? What addresses will be contained
in this packet? [6 marks]

8. What is the major difference between CSMA and CSMA/CD? What is the major difference between
CSMA/CD collision detection in wired LANs and wireless LANs? [2 marks]

Page 14 of 23
9. Figure 9 below represents routers with the indicated link costs in an institutional network. [5 marks]

~
,:.
2\'" 1
7
...
'•..•.,•..
<;i":''' ..... .

Figure 9

a. What is the name of the routing algorithm used to compute the following tables presented
in Figure 10 (below) for the least-cost paths in the institution's network?
b. Replace the blanks in Figure 10 (below) with the letters (Le., a,b,c) used to represent the
nodes in Figure 9 (above). E.g., "Node ~ table".
c. Use the arrows to show how the algorithm updates the tables in sequence. Draw the arrows
in Figure 10 (below).
d. Show how the costs in the circled row (i.e., "20 1") in Figure 10 (below) are derived. Present
your solution on page 16.
e. Do the tables in Figure 10 (below) require more updates? Justify your answer.

;----'.--------.-------.-~------~---.------"- --------~--.---~ .. -----_._,

cost to cost to cosito
Node_table a b c a b c a b c
a 0 2 7 a 0 2 3 a 0 2 3
E
.g b ~b 2 0 1
~ b 2 0 1
00 00 00

c 00 00 00 c 7 1 0 c 3 1 0

Node_table a
cog toc a b c
cost to cost to
i a b c
a 00 00 00 a 0 2 a 0 2 3
~ b 2 0 1
c
~b
0;;:::

c 7 1 0
.gE b 2 0 1
00 00 00 c 3 1 0

cost to costta cost to

a b c a b c a b c
:Node_table
a 00 00 00 a 0 2 7 a 0 2 3
~b 2 0 1 ~ cb 2 0 1
~ b 00 00 00 ...:
c 3 1 0
C 7 1 0 3 1 0
"." ..............•...,,"."".........................................." ........................................................................................." ..................................................- .......... time
Figure 10

Page 15 of 23
Page 16 of 23
10. Figures 12-16 presented on page 18 show relevant data for a TCP connection captured by Wireshark.
Figure 12 shows relevant packets captured. Figures 13-16 show associated data for each highlighted
packet. [6(3+3) marks]
a. What is the sequence number of the segment that is indicated by Wireshark as the No.4 packet
(Le., packet number 4)? Packet numbers are showed in the column that has the "No." as its
column header. Clearly indicate the position of this sequence number in the sender sequence
number space in Figure 11 below.

b. What is the packet number (indicated by Wireshark) of the segment that acknowledges the
receipt of the segment mentioned in point a (above)? What is the sequence number
acknowledged by this acknowledgement segment? Clearly indicate the position of this
sequence number in the sender sequence number space in Figure 11 below.

window size
14 IV -I

I111111111111111111111111
sender sequence number space

Figure 11

Page 17 of 23
No, TIme Source Destination Protocol Langtllnfo
,e'
,
1 ••ee.e 192.168.1.102 128.119:245.12 Tep 62 1161 ~ 80 [5¥N] Seq:8 Win=16384 Len--0 M55=1469 5ACKJEIv4=1
2 0.823172 128.119.245.12 192.168.1.102 Tep 62 80 • 1161 [S'ffl, ACK] Seq=0 Ack=1 Win=5840 Len-i HS5=1460 SACK}EPIoI=I
; 3 U23265 192.168.1.102 128.119.245.12 rcp 54 1161 • 89 [ACK] Seq:l Ack:l Win:17528 Len=8
i 4 0.826417 192.168.1.102 128.119.245.12 rep 619 [TCP segment of a reassembled POU]
, 50.841737 192.168.1.102 128.119.245.12 TCP 15,. [rCp segment of a reassembled POU]
I 6 0.053937 128.119.245.12 192.168.1.102 Tep 60 80 • 1161 [ACK] Seq:! Ack:566 Win=6780 Len=0
! 1 0.954026 192.168.1.182 128.119.245.12 TCP 15. [Tep segment of a reassembled POU]
i
8 0.054690 192.168.1.102 128.119.245.12 TCP 15. [TCP segment of a reassembled POU]
,
9 0.077294 128.119.245.12 192.168.1.102 rcp 60 80 • 1161 [ACK] Seq:! Ack=2026 Win=8760 Len:0
!
10 8.077405 192.168.1.102 128.119.245.12 rep 15m [Tep segment of a reassembled POU]
11 8.078157 192.168.1.102 128.119.245.12 TCP 15. [TCP segment of a reassembled POU]
12 •• 124085 128.119.245.12 192.168.1.102 rep 60 80 • 1161 !ACK] Seq:l Ack:3486 Win-11680 Len:0
Figure 12

on wire (4952 , 619 bytes bits)

Ethernet II, Src: ActionteJa:70:la (ee:20:e0:8a:70:la), Ost: LinksysG_da:af:73 (88:96:25:da:af:73)
Internet Protocol Version 4, Src: 192.168.1.102, Dst: 128.119.245.12
Transmission Control Protocol, Src Port: 1161 (1161), Dst Port: 80 (88), Seq: I, Ack: I, Len: 565
Figure 13

Figure 14

Frame 6: 60 bytes on wire (480 bits), 60 bytes captured (488

Ethernet II, Src: LinksysGJa:af:73 (08:06:25:da:af:73), Dst: ActionteJa:7B:la (ee:28:eB:8a:70: lal
Internet Protocol Version 4, Src: 128.119.245.12, Dst: 192.168.1.182
Transmission Control Protocol, Src Port: 80 (88), Ost Port: 1161 (1161), Seq: 1, Ack: 566, Len: 0
Figure 15

if Frame 7: 1514 bytes on , 1514

Ethernet II, Src: ActionteJa:78:1a lee:28:e0:8a:78:1al, Ost: linksysGJa:af:13 lee:86:25:da:af:73)
Internet Protocol Version 4, Src: 192.168.1.102, Dst: 128.119.245.12
Transllission Control Protocol, Src Port: 1161 (1161), Ost Port: ee lee),
Figure 16

Page 18 of 23
Section IV: Problem Solving [20 (10+10) Marks]

1. The Networking Team is evaluating the performance of a network. Alice is tasked to measure the
throughput of a TCP connection based on the Wireshark captured data presented on hardcopy. The
data Alice has received is presented in 9 figures. These figures are Figures 12-16 (presented on page
18) and Figures 17-20 (presented below).
As she examines the Wireshark captures, Alice believes that she would need more data to
complete this task. She believes that she has not received the complete captured data. The Team
Leader would like you to help Alice to identify the missing data so that she can measure the
throughput of the TCP connection. What other data is needed? Show your reasoning on page 20 (e.g.,
provide a formula with illustrations on how the throughput of a TCP connection can be measured).
[10 marks]

~ fraE 1514 bytes on 1514 bytes

~ Ethernet II, Src: ActionteJa:18:1a (":28:e8:8a:78:1a), Ost: LinksysGjla:af:73 (":86:25:da:af:13)
Internet Protocol Version 4, Src: 192.168.1.182, Ost: 128.U9.24S.12
TranSliission Control Protocol, Src Port: 1161 (11611, Ost Port: 88 {88" 3486, Ack: 1, Len: 1468
Figure 17

~ Fr. 9: 61 bytes on wire , 61 bytes captured

• Ethernet II, Src: LinksysG.da:af:73 (88:86:25:da:af:73l, Ost: ActionteJa:18:1a (88:28:e8:8a:71:1a)
~ Internet Protocol Version 4, Src: 128.119.245.12, Ost: 192.168.1.182
• Transmission Control Protocol, Src Port: 88 (88), Ost Port: 1161 (1161), I, Ack: 2826, Len: •
Figure 18

18: 1514 bytes on wire bits), 1514 bytes captured

• Ethernet II, Src: ActionteJa:71:1a 188:28:e8:8a:18:1a), Dst: LinksysGja:af:73 (88:86:25:da:af:13)
~ Internet Protocol Version 4, Src: 192.168.1.182, Ost: 128.119.245.12
Transmission Control Src Port: 1161 Ost Port: 88 Len: 1461
Figure 19

• Frae 12: 61 bytes on wire , 68 bytes captured

~ Ethernet II, Src: LinksysGjla:af:13 (88:86:25:da:af:73), Dst: Actionte_8a:7.:1a (88:28:e8:&a:18:1a)
• Internet Protocol Version 4, Src: 128.119.245.12, Ost: 192.168.1.182
~ Transmission Control Protocol, Src Port: 88 (88), Ost Port: 1161 (116ll, Seq: 1, Ack: 3486, Len: 8

Figure 20

Page 19 of 23
Page 20 of 23
2. A network designer is asked to address the reliable data transfer issues in a network. As you are
completing (5310, the network designer would like to seek your suggestions. Use the concepts and
techniques you have learned from (5310 to give your suggestions to the network designer.
Show steps whenever necessary. [10 marks]

Page 21 of 23
Extra Working Page 1

Page 22 of 23
Extra Working Page 2

THE END

Page 23 of 23
,,
,,
"