Examiner: RA. Moderator: RP. Subject Leader CB.
Semester 1 Examinations 2000
Duration: 2 Hours
Instructions to candidates
Do not open this question paper until you have been told to do so by the invigilator.
A figure in 
denotes the number of marks available for that question or part of
There are 4 questions. Answer 3 questions.
If you present answers to more than 3 questions, the first 3 of these answers will be
Questions carry 20 marks each.
otal number of marks available is 60.
Calculators may be used.
A printed language dictionary may be used.
The OSI 7
layer reference model describes a series abstractions for designing
computer networks beginning with the Physical Layer (layer 1) and
up to the Application Layer (layer 7).
With the aid of a diagram describe how a message is passed
between two network applications using the OSI 7
Discuss the reasoning behind the layering approach to network
Discuss a criti
cism you have of the OSI model and suggest
possible improvements to address this criticism.
Each host on the Internet is currently assigned an IP address 32 bits long
(IPv4). IPv4 addresses are usually written as a series of four decimal numbers
ween 0 and 255, e.g. 18.104.22.168. IPv4 addresses traditionally belong to
one of five classes of address, depending on the type of network.
Describe the structure of Class A, B and C addresses, and the types
of networks they are suited for.
What are the
advantages of subnetting? Propose a scheme for
subnetting a Class B network. Identify the mask to be used and the
maximum number of hosts supported by each subnet in the scheme.
The IEEE 802 series of standards describe both the Physical and Da
layers of their respective technologies. Two important standards are 802.3 and
802.5, respectively known as CSMA/CD (Ethernet) and Token Ring.
Describe three major differences between Token Ring and
Explain the mechanism for
collision detection and recovery in
Fast Ethernet and FDDI are both used to improve on ‘slow’ LAN technologies
such as 802.3/802.5.
Discuss the different approaches of the two technologies, assessing
them in terms of their applicability.
Explain with the aid of a diagram two methods to achieve fault
tolerance in FDDI, considering dual attach stations (DAS) and
single attach stations (SAS).
Consider a computer game server managing a networked game “ScaryBob”
with many players interact
ing via client machines. The need for players to
synchronise with each other requires the server to regularly update all clients’
clocks. Upon receipt of clock update the client acknowledges the update by
returning the time difference to the server. Outlin
e client and server functions,
based on the UDP protocol, to carry out this task. Explain how the functions
work and use pseudocode to show the processes required. (There is no need
for coding in any particular language, but include references to the socke
A company wishes to network its offices together to enhance the overall business
process. The company has three offices, one in London and two in Liverpool (next
door to each other). The London office is on three floors whilst the Liv
are both small single floor offices. The company wishes to be able to handle a range
of data from email and documents to video conferencing. Finally the company wishes
to connect to the external Internet. Much data on the company’s computer
highly sensitive and the company is concerned about the security issues of the new
network, both internally and externally.
Specify, with justification, a network that will meet these requirements. You
may ignore any application software that i
How might the company expand their network to cope with a rapid expansion
of their London office, both in terms of data throughput and number of nodes?
The Internet is based upon the TCP/IP protocol suite with further protocols
rating either at the same layers as TCP/IP or above. Compare TCP and
UDP, giving two advantages of each and suggest an application using each.
Justify your answers.
Data travelling on a network may either be Packet
Switched or Circuit
Switched data may be transported using either Datagrams
(connectionless) or Virtual
oriented). Explain the
differences between the switching and service types, and the major application
Asynchronous Transfer Mode (
ATM) is a high
speed network technology.
ATM has a physical layer, an ATM layer and an ATM Adaptation Layer
(AAL). Four protocols at the AAL layer allow for the creation of a variety of
What are the factors that determine the service clas
ses and what is
the purpose of such distinction?
Discuss how AAL 5 is considered more appropriate for Internet
services, by comparing it to the other adaptation layer protocols
Demand (VOD) server stores and distributes MPEG en
films. Each film is 2 hours long and contains 3.5Gigabytes of data. The server
is connected to an ATM network. Calculate the required bandwidth (in
Megabits per second) of the ATM link required to support 50 different streams
There are no corrections on this proof copy and the examination paper is now ready for printing. I am aware
that the printed papers will be produced directly from this proof.
Special printing instructions
for the examination paper
Signed .................................................................... Date ...............7
STATIONERY AND OTHER REQUIREMENTS
14 PAGE ANSWER BOOKS
6 PAGE ANSWER BOOKS
DRAWING PAPER A1
DRAWING PAPER A2
DRAWING PAPER A3
TRACING PAPER A1
TRACING PAPER A2
TRACING PAPER A3
ANY OTHER REQUIREMENTS
Are there any special items which may be taken into the examination by candidates? If so, please list here