Important RGPV Question
CS-802 (C) High Performance Computing
VIII Sem, CSE
UNIT 1-Introduction To Modern Processors
Q.1) Draw the block diagram of a typical cache-based microprocessor.
(RGPV May 2022)
Q.2) Differentiate Multi-core, Multithreaded and Vector processors.
(RGPV May 2022)
Q.3) Discuss C++ optimizations.
(RGPV May 2022)
Q.4) What do you mean by scalar profiling? Explain function and line-based profiling.
(RGPV May 2022)
Q.5) Explain the classification of Clusters in detail.
Q.6) Examine the concept of multiprogramming.
Q.7) Describe Cluster Systems and its components. Explain its architecture.
Q.8) Explain various clusters classifications along with their applications.
UNIT 2-Data Access Optimizations
Q.1) What is Parallelism? Discuss the different variants of parallelism with the help of an example in each case.
(RGPV May 2022)
Q.2) Discuss the performance problems that can be raised with OpenMP shared-memory programming, also discuss how they can be circumvented?
(RGPV May 2022)
Q.3) What is meant by parallel programming?
Q.4) Give the issues in parallel and distributed paradigms.
Q.5) Discuss the importance of High-Performance Computing.
UNIT 3-Basics Of Parallel Computing
Q.1) What is OpenMP? Explain the concepts of data scoping, work-sharing for loops and loop scheduling.
(RGPV May 2022)
Q.2) Define concept of multiprogramming.
Q.3) Discuss parallel computing with suitable examples.
Q.4) List the two dominating programming language groups for HPC.
UNIT 4-Efficient Open MP Programming
Q.1) Discuss the significance of program profiling in identifying performance bottlenecks in OpenMP applications.
Q.2) Describe common tools or methods used for profiling parallel OpenMP codes.
Q.3) Elaborate on the concept of ‘load imbalance’ as a performance pitfall in OpenMP. Provide an example of how it can arise and suggest strategies to mitigate it.
Q.4) Explain why excessive synchronization can degrade performance in OpenMP programs. Illustrate with an example of a common synchronization construct that might lead to this.
Q.5) What is the “granularity” of a parallel task, and why is it a critical factor in determining whether to parallelize a short loop?
Q.6) Define ‘serialization’ in the context of OpenMP performance. Provide an example of a common programming construct or pattern that can introduce serialization.
UNIT 5-Distributed Memory Parallel Programming With MPI
Q.1) Explain the blocking and rolling strategies with the help of multiplication of a sparse matrix with a vector.
(RGPV May 2022)
Q.2) What is non-blocking point-to-point communication? How it differs from blocking communication?
(RGPV May 2022)
Q.3) How to reduce the communication overhead? Differentiate non-blocking and asynchronous communication.
(RGPV May 2022)
Q.4) Draw the flowchart for distributed-memory parallelization of the Jacobi algorithm.
(RGPV May 2022)
Q.5) Differentiate between PVM and MPI.
Extra Questions-
Q.1) Write a short note on the following (any two)-
a) Moore’s law
b) Scalability Laws
c) Bandwidth-based performance modelling
d) Virtual topologies
(RGPV May 2022)
— Best of Luck for Exam —
