Gujarat Technological University 2010 M.E Distributed operating system - Question Paper

Seat No.    Enrolment No.

GUJARAT TECHNOLOGICAL UNIVERSITY

M.E Sem-II Examination July 2010

Subject code: 720201 Subject Name: Distributed Operating System Date: 05 /07 /2010    Time: 11.00am - 1.30pm

Total Marks: 60

Instructions:

1.    Attempt all questions.

2.    Make suitable assumptions wherever necessary.

3.    Figures to the right indicate full marks.

Explain distributed computing model.    06

1.    Increasing replica in a Distributed operating system will improve the overall 02 performance of the system? Justify your answer with relevant example.

2.    Explain polling and Interrupt in message passing and explain any one    04 algorithm for the same.

Design a four way hand shaking protocol for message passing that reduce the 06 communication overhead and increase performance compare to four way handshake message passing protocol in network operating system.

Explain desirable features of Process migration in Distributed Operating system 06

OR

Design a distributed algorithm for multiple producers and multiple consumer 06 problems that have to be implemented in network operating system environment to achieve transparency and consistency like Distributed operating system.

Explain Distributed clock synchronization algorithm.    06

Explain the requirement of cooperative load sharing approach or load balancing 06 approach? Give an example that specifically requires cooperative approach?

OR

Explain distributed algorithm for Deadlock detection and prevention    06

1.    What is false sharing in a DSM system? When is it likely to occur?    03

2.    What are the main causes of thrashing in a DSM system?    03

Explain Strict Consistency model? If we want to implement the model in real 06 life what modifications are required? Make suitable assumption.

What trend are modern distributed file systems following for kernel design? 06 What are the main reasons for using this approach?

OR

1.    Most DSM systems in which caching is managed by the operating system, 03 use the write invalidate scheme for consistency instead of the write-update scheme. Explain why.

2.    Differentiate between PRAM consistency and processor consistency.    03 Describe the process model of Amoeba. 06

1. In what aspects is the design of a distributed file system different from that 04 of a file system for a centralized time sharing system?

2. When file system replicates files, they do not replicate all files. Give an 02 example of a kind of file that is not worth replicating.

1.    Discuss the relative advantages and disadvantages of using full-file caching 03 and block caching models for the data-caching mechanism of a distributed file system.

2.    In the design of distributed file system, high performance and high 03 reliability are conflicting properties. Discuss.

OR

1.    Discuss disadvantages of stateful file servers. Give an example, where it 03 might be necessary to use stateful file servers.

2.    Replication and caching - both concepts are not required in the same 03 distributed system. Is the statement True or False? Justify your answer.

1.    When session semantics are used, it is always true that changes to a file are 02 immediately visible to the process making the change and never visible to processes on other machines. Is the statement True or False? Justify your answer.

2.    In the design of distributed file system, high availability and high scalability 04 are mutually related properties. Discuss.

*************

2

Attachment: gujarat-technological-university-2010-m-e--52861-1.pdf

Earning:   Approval pending.