GTU Parallel Processing - June 2015 Exam Question Paper

GTU Computer Engineering (Semester 8)
Parallel Processing
June 2015

Total marks: --
Total time: --

INSTRUCTIONS
(1) Assume appropriate data and state your reasons
(2) Marks are given to the right of every question
(3) Draw neat diagrams wherever necessary

1 (a) Define Latency and Bandwidth of memory. Briefly explain different ways to minimized latency and to increase bandwidth of memory?

7 M

1 (b) Briefly explain different classification of parallel computers

7 M

2 (a) Briefly explain static and dynamic interconnection networks for parallel Computers.

7 M

2 (b) Define following terms and explain the importance related to parallel algorithm design
1. Decomposition.
2. Concurrency.
3. Granularity.

7 M

2 (c) Briefly explain usefulness of task dependency and task interaction graph related to parallel algorithm design. Draw task interaction graph for sparse matrix vector multiplication.

7 M

3 (a) Briefly explain following decomposition techniques used in parallel algorithm design
1. Data decomposition
2. Exploratory decomposition
3. Speculative decomposition.

7 M

3 (b) Briefly explain one to all broadcast and all to one reduction on eight node hypercube. Also find the cost of communication for one to all broadcast on eight node hypercube.

7 M

3 (c) Briefly explain all to all personalized communication and its applications. Briefly explain an optimal algorithm of all to all personalized communication on eight node hypercube.

7 M

3 (d) Briefly explain loop splitting, self scheduling and chunk scheduling for task mapping to achieve load balancing among processes.

7 M

4 (a) Enlist various performance metrics for parallel systems. Explain Speedup, Efficiency and Cost in brief.

7 M

4 (b) Define Isoefficiency function and derive equation of it.

7 M

4 (c) Briefly explain the relation between Speedup and efficiency as functions of the number of processing elements. Derive the equation which relate speedup and efficiency with processing elements.

7 M

4 (d) Briefly explain Four different implementation of Send and Receive operations. Briefly explain Send and receive functions of MPI .

7 M

5 (a) Explain following MPI routines with arguments.
MPI_Gather.
MPI_Scatter.
MPI_Reduce.

7 M

5 (b) Briefly explain Cannon's algorithm for Matrix-Matrix multiplication. What are the advantages of this algorithm over other parallel algorithm of matrix Multiplication?

7 M

5 (c) Briefly explain different synchronization primitives available in Pthread. Explain three types of mutex (normal, recursive and error check) in context to Pthread.

7 M

5 (d) Briefly explain parallel algorithm of Quick sort with example for shared address space parallel computer.

7 M

More question papers from Parallel Processing

SPONSORED ADVERTISEMENTS