BE CSE 411603-611602 Distributed Computing November 2010 Sathyabama University

Register Number

                               

SATHYABAMA UNIVERSITY

(Established under section 3 of UGC Act, 1956)

Course & Branch: B.E-CSE

Title of the Paper: Distributed Computing            Max. Marks: 80

Sub. Code: 411603-611602                                    Time: 3 Hours

Date: 23/11/2010                                                    Session: AN

______________________________________________________________________________________________________________________

PART - A                (10 X 2 = 20)

Answer ALL the Questions

1.     Draw the processor pool model and explain its advantages.

2.     How does the redundancy technique help in implementing the fault tolerance concept?

3.     Differentiate the shared data approach and the message passing approach.

4.     How can the buffer overflow problem be handled?

5.     Define thrashing. When does it occur?

6.     Highlight the advantages of using the Distributed Shared memory.

7.     Differentiate between internal synchronization and external synchronization of clock in a distributed system.

8.     List the advantages and disadvantages of process migration.

9.     What is meant by atomic transaction?

10.   Define replication transparency.

   PART – B                    (5 x 12 = 60)

Answer All the Questions

11.   Explain the various transparency issues involved in designing a distributed operating system.

(or)

12.   Suppose a component of a distributed system suddenly crashes. How will this event inconvenience the user when:

        (a) The system uses the processor-pool model and the crashed component is a processor in the pool.

        (b) The system uses the processor-pool model and the crashed component is a user terminal.

        (c) The system uses the workstation-server model and the crashed component is a server machine.

        (d) The system uses the workstation-server model and the crashed component is a user workstation.

       

13.   (a) Explain what is meant by absolute ordering, consistent ordering and causal ordering of messages. Give a mechanism to implement each one.

        (b) The operation performed by a server are idempotent. Describe a mechanism for implementing exactly – once semantics in this case.

(or)

14.   (a) What is “piggybacking” of a message? How does it help in reducing network traffic? Give some examples of where the piggybacking scheme may be used to improve performance in distributed systems.

        (b) Explain in detail how is the group communication implemented?

15.   Explain the various consistency models in detail.

(or)

16.   Explain in detail about the heterogeneous distributed shared memory concept.

17.   Define mutual exclusion. Explain the various methods adopted to implement the mutual exclusion property.

(or)

18.   Explain in detail the various issues in designing load balancing algorithms.

19.   Explain in detail about the File caching and the cache validation schemes.

(or)

20.   Define fault tolerance. Explain in detail about the stateful file servers and the stateless file servers. Also explain about the advantages of stateless service paradigm in crash recovery.

Share This