DEC 1997

Question Paper of CS-13 – Operating Systems of Dec 1997 from IGNOU

Time : 3 Hours
Max. Marks : 75

Note: Question 1 is compulsory.
Answer any three from the rest.

1.
(a) Assume that in the system shown in the following figure, Process p1 does not won a disk drive and it requests two disk drives simultaneously. Illustrate that situation by means of the general resource graph and use the deadlock detection algorithm to evaluate the resulting system state.

(b) Discuss whether each of the following programming techniques and program action is good or bad with regard to the degree pf locality of page reference it is likely to exhibit. Explain your reasoning and where applicable, state roughly the number of distinct loci of activity (hot spots) that you expect the execution to generate.

(i) Sequential processing of one dimensional array
(ii) Sequential processing of two dimensional array
(iii) Hashing
(iv) Interrupt servicing
(v) Indirect addressing
(vi) Procedure invocation

(c) Discuss why shared bus multiprocessors are generally regarded as having limited scalability.

2.
(a) Write a program/algorithm that solves the readers/writer problems by using monitors and also explain it.

(b) Compare and contrast the semaphore to the same problem [2(a)] in terms of the type of data abstraction and readability of the code.

3.
(a) When do page-faults occur ? Describe the action taken by the O.S when page fault occurs.

(b) Describe what is Belady’s anomaly and provide an example that illustrates anomalous behaviour of FIFO.

4. Discuss RSA algorithm (related to cryptography) and explain its working through one example.

5.
(a) Point out and discuss the major differences in resource management system requirement between uniprocessor and multiprocessor O.S.

(b) Discuss the operation of multistage switch-based system.

6. Provide a detailed step by step comparison of Lamport’s and Ricart and Agrawala’s algorithm for mutual exclusion. Identify key differences and explain where the saving in the number of messages required by the latter algorithm come from ? Assess and compare the difference, if any, in the typical duration of unavailability of the target resource due to synchronization caused by each algorithm.

Comments are closed.