Pune University BE (Computer Engineering) Advanced Computer Architecture Question Papers

B.E. (Computer Engg.) ADVANCED COMPUTER ARCHITECTURE (2008 Pattern) (Sem. – II)

Time :3 Hours]                                                                                              [Max. Marks :100

Instructions to the candidates :-

1)            Attempt: Q. 1 or Q. 2, Q. 3 or Q. 4, Q. 5 or Q. 6 from section I.

Attempt: Q. 7 or Q. 8, Q. 9 or Q. 10, Q. 11 or Q. 12 from section II.

2)            Answers to the two sections should be written in separate answer books.

3)            Neat diagrams must be drawn wherever necessary.

4)            Pigures to the right indicate full marks.

5)            Assume suitable data, if necessary.

SECTION – I

QIA a) What is significance of ‘Scalability? Derive the Amdahl’s law for speedup performance.                                                                                                             [10]

b) Discuss Feng’s classification for parallel computer Architecture in detail.           [8]

OR

Q2) a) Explain the important Architecture features of explicitly parallel instruction computing. (EPIC)                                                                                             [10]

b) Explain with suitable example – various levels of pipelining.                         [8]

Q3) a) Discuss various pipeline hazards. Give hazard detection and resolution techniques.        [8]

b) Explain internal forwarding with proper example.                                            [8]

OR

Q4) a) Prove that ‘n’ stage pipeline processor can be at most ‘n’ times faster than the corresponding non-pipelined processor.                                                        [8]

b) Explain any four features of ultra SPARC. Explain the concept of Register Stack Engine (RSE).                                                                                                           [8]

 

With suitable example explain following features implemented in CRAY-I architecture            [8]

i)            

Vector chaining.

ii)           Vector looping.

With suitable example, explain the necessity of data Routing in Array Processors.           [8]

OR

Discuss a problem of 3*3 matrix multiplication on mesh network. Obtain it’s time complexity.                                                                                                                                [8]

Discuss any one parallel sorting algorithm for Array processor. [8]

SECTION – II

Explain the architecture of IBM Power 4 Processor.                                    [8]

What is Multiport Memories? Explain in brief.                                             [8]

OR

Compare loosely coupled and Tightly coupled multiprocessor system. [8] Compare between:   [8]

i)              Write – Thought and Write – Back caches.

ii)           Write – Update and Write – Invalidate protocol.

Compare and explain various multithreaded processor architecture. [8] Explain with suitable examples shared memory parallel programming. [8]

OR

Discuss the various contest switching policies implemented in multithreading architecture.      [8]

Explain with suitable example message passing parallel programming. [8]

What are the issues in multiprocessor Operating system? Discuss in detail.           [10]

Write note on Neuro-Computing paradigms.                                                 [8]

OR

State and explain control and data parallelism used in CCC by means of standard constructs.    [8]

Write parallel algorithm to computer matrix.                                             [10]

Z = AB + (C+D)*(I+G)

Leave a Comment