BSc (Hons) Computer Network with Security

BSc (Hons) Computing with Information

Systems
Cohort: BCNS/11/FT, BCIS/12/FT(Top Up), BCIS/11/PT

Examinations for 2012- 2013 Semester 2

MODULE: Distributed Systems

MODULE CODE: SCG 3102C/SCG 3125

Duration: 3 Hours

Instructions to Candidates:

1. Answer all questions.

2. Questions may be answered in any order but your answers must show
the question number clearly.
3. All questions carry equal marks of 20
4. Total marks 100.
5. Calculators are allowed.

This question paper contains 4 questions and 7 pages.

Page 1 of 7
Distributed Systems (SCG3102C/SCG3125) SITE/DA 2012-2013 S2
QUESTION 1: (20 MARKS) Characteristics of DSs

(a) List and describe any 4 characteristics of distributed systems.

(4 marks)

(b) State the criterion for a system to be considered as not a distributed one.

(3 marks)

(c) Give 2 types of hardware resource and 2 types of data or software

resource that can usefully be shared as it occurs in distributed systems.

(4 marks)

(d) Explain the evolution from mainframe to cloud computing.

(4 marks)
(e) Imagine a client server inter process communication.

(i) Explain the use of port and socket.

(ii) Why is it important not to leave a port open? Relate any

incident which happened in the past decade.

(5 marks)

Page 2 of 7
Distributed Systems (SCG3102C/SCG3125) SITE/DA 2012-2013 S2
QUESTION 2: (20 MARKS) RMI

(a) Figure 2 illustrates an RMI communication.

Figure 2

Describe the role of each component in the RMI.

(8 marks)

(b) Explain the terms marshalling and un-marshalling as per an RMI or RPC.

(4 marks)

(c) Compare RMI and CORBA from a middleware perspective.

(4 marks)

(d) What are purpose of

(i) a remote interface for a remote object?

(ii) instantiation of remote objects?

(4 marks)

Page 3 of 7
Distributed Systems (SCG3102C/SCG3125) SITE/DA 2012-2013 S2
QUESTION 3: (20 marks) Name Services

(a) Consider the URL in the table below.

http://www.utm.ac.mu:8888/SiteProgrammes/BCIS.pdf

Illustrate with help of a diagram the results of a DNS lookup and an ARP lookup
assuming we are using IPv4.
(5 marks)

(b) Why do DNS root servers hold entries for two-level names such as ac.mu and
un.org, rather than one-level names such as .mu and .org?

(4 marks)

(c) Is it possible for a DNS server to provide multiple answers to a single name
lookup? When and Why?
(5 marks)

(d) Explain the use

(i) of aliases in name services

(ii) of caching servers to enhance availability

(6 marks)

Page 4 of 7
Distributed Systems (SCG3102C/SCG3125) SITE/DA 2012-2013 S2
QUESTION 4: (20 MARKS) Time Synchronization

(a) Why is computer clock synchronization of paramount importance? Give 2

business scenarios whereby time accuracy is critical?

(4 marks)

(b) Computers clocks can be synchronized through peers, UTC and NTP
servers.

(i) How might the clocks in two computers that are linked by a local
network be synchronized without reference to an external time
source?

(ii) What factors limit the accuracy of the procedure you have
described in (i)?

(iii) How could the clocks in a large number of computers connected

by the Internet be synchronized? Discuss the accuracy of that
procedure.

(6 marks)

(c) Logical time is more efficient for tracking processes in a distributed system
and to model the time drift and skew. Explain how.

(4 marks)

Page 5 of 7
Distributed Systems (SCG3102C/SCG3125) SITE/DA 2012-2013 S2
 (d) A client attempts to synchronize with a time server. It records the round-
trip times and timestamps returned by the server in the table below.

(i) Which of these times should it use to set its clock? To what time
should it set it?

(ii) Estimate the accuracy of the setting with respect to the server’s
clock.

(iii) If it is known that the time between sending and receiving a

message in the system concerned is at least 8 ms, do your
answers change?

(6 marks)

Page 6 of 7
Distributed Systems (SCG3102C/SCG3125) SITE/DA 2012-2013 S2
QUESTION 5: (20 MARKS) Fault Tolerance

(a) Three computers together provide a replicated service. The manufacturers

claim that each computer has a mean time between failure of five days; a
failure typically takes four hours to fix. What is the availability of the
replicated service?

(5 marks)

(b) In the gossip architecture, why does a replica manager need to keep both
a ‘replica’ timestamp and a ‘value’ timestamp?

(5 marks)

(c) Explain the difference between linearizability and sequential consistency,

and why the latter is more practical to implement, in general.

(5 marks)
(d) Explain why allowing backups to process read operations leads to
sequentially consistent rather than linearizable executions in a passive
replication system.

(5 marks)

***END OF QUESTION PAPER***

Page 7 of 7
Distributed Systems (SCG3102C/SCG3125) SITE/DA 2012-2013 S2