Finite-Element Sparse Matrix Vector Multiplication on Graphic Processing Units

Author

Dehnavi, Maryam Mehri ; Fernández, David M. ; Giannacopoulos, Dennis

Author_Institution

Dept. of Electr. & Comput. Eng., McGill Univ., Montreal, QC, Canada

Volume

46

Issue

8

fYear

2010

Firstpage

2982

Lastpage

2985

Abstract

A wide class of finite-element (FE) electromagnetic applications requires computing very large sparse matrix vector multiplications (SMVM). Due to the sparsity pattern and size of the matrices, solvers can run relatively slowly. The rapid evolution of graphic processing units (GPUs) in performance, architecture, and programmability make them very attractive platforms for accelerating computationally intensive kernels such as SMVM. This work presents a new algorithm to accelerate the performance of the SMVM kernel on graphic processing units.

Keywords

computer graphic equipment; coprocessors; finite element analysis; matrix multiplication; sparse matrices; finite-element electromagnetic applications; finite-element sparse matrix vector multiplication; graphic processing units; sparsity pattern; Acceleration; Computer architecture; Finite element methods; Graphics; Iron; Iterative methods; Kernel; Parallel processing; Sparse matrices; Yarn; Computer architecture; graphic processing units (GPUs); parallel processing; sparse matrix vector multiplication (SMVM);

fLanguage

English

Journal_Title

Magnetics, IEEE Transactions on

Publisher

ieee

ISSN

0018-9464

Type

jour

DOI

10.1109/TMAG.2010.2043511

Filename

5512911