CIS222 - Data Communications and Enterprise Networking

advertisement

UNIVERSITY OF LONDON

GOLDSMITHS COLLEGE

B. Sc. Examination 2010/11

COMPUTING AND INFORMATION SYSTEMS

INFORMATION TECHNOLOGY

IS52016A

Data Communications and Enterprise Networking

Duration: 3 hours

Date and time:

This paper is in two parts, Part A and Part B. There are a total of three questions in each part. You should answer two questions from Part A and two questions from

Part B.

Full marks will be awarded for complete answers to a total of four questions, two from

Part A and two from Part B. Each question carries 25 marks. The marks for each part of a question are indicated at the end of the part in [.] brackets.

There are 100 marks available on this paper.

No calculator should be used.

THIS PAPER MUST NOT BE REMOVED FROM THE EXAMINATION

ROOM

IS52016A 2010/11 1 TURN OVER

PART A

Question 1

(a) State, in your answer book, which two of the following statements are true and which two are false. For each of the false statements, write out a corrected true statement: i.

Differential Manchester coding represents a 1 by a transition at the start of the bit, a 0 by no transition at the start of the bit and has an extra transition in the middle of the bit. ii.

Delay distortion, otherwise known as inter-symbol interference, is caused by different frequencies travelling at different speeds in the medium. iii.

Echo cancellation is a technique that eliminates echoes by making a full duplex line half duplex. iv.

FDDI uses the same format of addresses as Ethernet and Token Ring networks.

[3]

(b) Write down the Shannon’s Law equation, explaining each of the terms and giving the units in which they are usually measured. [3]

Use this equation to derive an expression for the capacity provided by a channel with a bandwidth of 40 MHz when the channel is subject to a signal to

[3] noise ratio of 30dB.

(c) Describe the difference between the Ethernet II and IEEE802.3 frame structures. Describe how each protocol demultiplexes to the network layer and how both protocols can co-exist on the same LAN.

[4]

(d) Name and describe an access method used on some LANs that can guarantee that frames can always be transmitted within a specified timeframe. [4]

(e) Show how a CRC-3 checksum can be calculated to provide error detection for the byte 10100111 using the generator 1001. [3]

Suppose that a byte followed by a redundant CRC-3 checksum was received as 01001110110. Carry out a CRC check using the generator 1001 and state whether or not an error has occurred. [5]

IS52016A 2010/11 2 TURN OVER

Question 2

(a) State, in your answer book, which two of the following statements are true and which two are false. For each of the false statements, write out a corrected true statement: i.

The ICMP time exceeded message is generated whenever an IP datagram is discarded because it has not reached its final destination within 500ms. ii.

IP version 6 incorporates IPSec as its network layer security mechanism. iii.

At each end of a TCP connection, an initial sequence number is generated derived from their system clocks. iv.

If an error is detected in the TCP checksum, a negative acknowledgment is sent to the transmitter. [3]

(b) An application passes 2,048 bytes of data to UDP which after adding its 8-byte header passes it onto IP, which needs to transmit it onto an Ethernet LAN with an MTU size of 1500 bytes. Describe how IP will handle this request at the transmitting host (assuming a standard IP header of 20 bytes) and at the receiving host, paying particular attention to packet sizes, fragment offsets and the use of the MF bit.

(c) Describe how a TCP connection is identified and how this identifier can be

[6]

[2] guaranteed to be unique.

(d) What is meant by classful and classless routing? [4]

Describe why it was necessary to introduce Classless Internet Domain Routing on the Internet. [3]

(e) An ISP is able to allocate host addresses from the range 135.47.128.1 to

135.47.191.254. What is the network address of the ISP (including the network prefix) and what network mask should it use? What is the maximum number of host addresses that it could allocate? [4]

The ISP decides to split its whole address range into a number of /23 networks for its customers. How many such network addresses can it allocate and how many hosts can then be allocated in each /23 network? What are the first two

/23 network addresses that the ISP could allocate to its customers? You can assume that you can use the network with all 0s. [3]

IS52016A 2010/11 3 TURN OVER

Question 3

(a) State, in your answer book, which two of the following statements are true and which two are false. For each of the false statements, write out a corrected true statement: i.

Trivial File Transfer Protocol was designed to transmit very short files over

Wide Area Networks. ii.

Client Server protocols are best implemented using UDP rather than TCP. iii.

At the top level of the Domain Name System are thirteen replicated root servers which know the authoritative DNS servers for each of the top level domains on the Internet. iv.

Transport Protocol Data Units are passed from the Application Layer to the

Transport Layer. [3]

(b) Describe three advantages of using a layered architecture for networking. [3]

Give three examples of models that use layered architectures. [3]

(c) Explain why the MIME protocol was developed to work alongside SMTP. [5]

(d) Describe the hierarchical structure of a typical domain name using an example and show how the allocation of names is delegated at four different levels. [4]

(e) Show how the byte 11001101 can be coded using an even Hamming Code to support single bit error detection and correction. [3]

Another even Hamming coded byte was received with one bit corrupted and the bits received were 101101000101. Show how the error can be detected and then corrected. What was the byte that was encoded originally? [4]

IS52016A 2010/11 4 TURN OVER

PART B

Question 4

(a) State, in your answer book, which two of the following statements are true and which two are false. For each of the false statements, write out a corrected true statement: i.

A universal service obligation requires a network operator to deliver all of its services anywhere in the world. ii.

USB is a bit-oriented master/slave protocol which uses CRC error detection. iii.

Dark fibre is a type of optical fibre that carries light signals just outside the spectrum of visible light. iv.

Wireless LANs make use of the Carrier Sense Multiple Access/Collision

Avoidance access method.

[3]

(b) Describe the three generic strategies (cost leadership, differentiation and focus strategies) that an enterprise can employ to gain competitive advantage. [6]

(c) Describe, with the aid of a diagram, how an FDDI ring can recover from a single cable break. [5]

(d) Describe, with the aid of an example, what each of the three parts of a standard Ethernet designation defines.

(e) Use the Spanning Tree Protocol to determine which bridge ports should be

[4] blocked in the following LAN topology. To do this, draw this diagram in your answer book and then show which bridge is elected as the root bridge by means of a thick lined box and show the path costs from each bridge port to the root bridge. Mark all the root ports with an R and all the designated ports with a D and all the blocked ports with an X. Draw the spanning tree with thick lines on the diagram. [7]

100Base-T2

Cost = 19

Priority 2

Address 1

Priority 1

Address 3

Priority 2

Address 2

10Base-T

Cost = 100

Priority 2

Address 5

10Base-T

Cost = 100

Priority 1

Address 4

1000Base-T2

Cost = 4

IS52016A 2010/11 5 TURN OVER

Question 5

(a) State, in your answer book, which two of the following statements are true and which two are false. For each of the false statements, write out a corrected true statement: i.

The North American SONET and European SDH transmission networks have common speeds and frame structures at 155Mbit/s and above. ii.

Dual Tone Multiple Frequency signalling is used between telephone exchanges to set up calls. iii.

GSM phones use a combination of frequency and code division multiplexing to communicate with base stations. iv.

Frame Relay networks can use the E.164 international telephone numbering scheme for addressing. [3]

(b) Describe the process that takes place when a call is made from a mobile phone to a land-line number.

(c) Describe how an IP VPN can be set up over the Internet.

[5]

Why are IP VPNs popular with businesses?

[3]

[2]

(d) What is meant by administrative distance and when does a router take account of it?

(e)

Draw the network diagram below in your answer book and use Dijkstra’s

[4] algorithm to calculate the shortest route between A and E, where the numbers represent distances between the nodes. On your diagram, show the node labels you have used at each step of the algorithm and mark the shortest path with a thick line.

A

[6]

H

7

1

2

B

12 11

G

7

F

13

10

6

3

4

9

D

A

5

C

E

In your answer book, draw up a routing table which shows the next hop and the shortest distance from node A to each of the other nodes on the network.

[2]

IS52016A 2010/11 6 TURN OVER

Question 6

(a) State, in your answer book, which two of the following statements are true and which two are false. For each of the false statements, write out a corrected true statement: i.

Frame Relay is a connectionless protocol that can operate at up to 2 Mbit/s. ii.

Fully meshed networks provide good resilience and scalability. iii.

Remote Monitors can be embedded in LAN switches to allow networks to be monitored remotely. iv.

SNMP uses UDP as its transport layer. [3]

(b) Describe three ways in which a network design can be tested. [3]

(c) Briefly describe the five network management functions in the ISO

Management Framework. [5]

(d) List seven items of information that you would expect to see in a trouble ticket. [7]

(e) What is meant by availability? How is it related to the Mean Time to Repair

[3] and the Mean Time Between Failures?

Two LANs are connected by two private circuits which are diverse and separate. The two circuits are each connected via a different router on each

LAN. The availability of each circuit is 99.8% and the availability of each router is 99.9%. Derive an expression for the overall availability of communications between the two LANs. [4]

IS52016A 2010/11 7 END OF EXAMINATION

Download