Study of Parallelization of Algorithms using OpenMP

This book provides a set of pragma's for programmers to parallelize their code. It explains the concept of parallel processing and the concept of OpenMP and how it is used in the C/C++ programs for effective utilization of available no. of cores/processors. Also illustrates the basic concepts of parallel computing with a brief overview of OpenMP. Describes an analysis of algorithms from different fields like Matrix Multiplication, Saddle point, Cholesky decomposition, LU decomposition, Pi Value using Monte Carlo method. The observations and results obtained show that how the usage of OpenMP's Pragma are effective in the normal C/C++ programs and how the result varies according to the inputs and available no. of threads and shows that it is useful only when we are working on large data set or large computations are involved in the given problem.

Autorentext

Prakash Choudhary travaille en tant que professeur adjoint au NIT Manipur. Il a obtenu son doctorat au MNIT Jaipur en 2016. Il est diplômé du VNIT Nagpur en 2012 et de l'Université du Rajasthan en 2009. Son domaine de recherche actuel est les systèmes distribués, la reconnaissance des formes et le traitement des images.

Klappentext

This book provides a set of pragma's for programmers to parallelize their code. It explains the concept of parallel processing and the concept of OpenMP and how it is used in the C/C++ programs for effective utilization of available no. of cores/processors. Also illustrates the basic concepts of parallel computing with a brief overview of OpenMP. Describes an analysis of algorithms from different fields like Matrix Multiplication, Saddle point, Cholesky decomposition, LU decomposition, Pi Value using Monte Carlo method. The observations and results obtained show that how the usage of OpenMP s Pragma are effective in the normal C/C++ programs and how the result varies according to the inputs and available no. of threads and shows that it is useful only when we are working on large data set or large computations are involved in the given problem.