Matrix Multiplication in Multiphysics Systems Using CUDA

Document Type

Conference Proceeding

Publication Date

2014

Publication Title

Proceedings of the 18th International Conference on Systems Science (ICSS 2013), Advances in Intelligent Systems and Computing

Publisher

Springer Verlag

Volume

240

First page number:

493

Last page number:

502

Abstract

Multiphysics systems are used to simulate various physics phenomena given byPartial Differential Equations (PDEs). The most popular method of solving PDEs is Finite Element method. The simulations require large amount of computational power, that is mostly caused by extensive processing of matrices. The high computational requirements have led recently to parallelization of algorithms and to utilization of Graphic Processing Units (GPUs). To take advantage of GPUs, one of GPU programming models has to be used. In this paper, CUDA model developed by nVidia is used to implement two parallel matrix multiplication algorithms. To evaluate the effectiveness of these algorithms, several experiments have been performed. Results have been compared with results obtained by classic Central Processing Unit (CPU) matrix multiplication algorithm. The comparison shows that matrix multiplication on GPU significantly outperforms classic CPU approach.

Keywords

CUDA; libMesh; Matrix Multiplication; Multiphysics Simulations

Disciplines

Electrical and Computer Engineering

Language

English

Permissions

Use Find in Your Library, contact the author, or interlibrary loan to garner a copy of the item. Publisher policy does not allow archiving the final published version. If a post-print (author's peer-reviewed manuscript) is allowed and available, or publisher policy changes, the item will be deposited.

UNLV article access

Search your library

Share

COinS