An Improved Parallel Algorithm for Computing Approximate Inverses by Reducing Synchronizations

Author

Gravvanis, George A. ; Giannoutakis, Konstantinos M.

Author_Institution

Dept. of Electr. & Comput. Eng., Democritus Univ. of Thrace, Xanthi, Greece

fYear

2008

fDate

1-5 July 2008

Firstpage

237

Lastpage

243

Abstract

A new parallel algorithm, based on the concept of anti diagonal wave pattern, for computing approximate inverses, is introduced for symmetric multiprocessor systems. The parallel normalized approximate inverses are used in conjunction with parallel normalized preconditioned conjugate gradient-type schemes, for the efficient solution of sparse finite element linear systems. The parallel implementation issues of the new algorithm are discussed and the parallel performance is presented, using OpenMP.

Keywords

approximation theory; conjugate gradient methods; finite element analysis; linear systems; mathematics computing; matrix inversion; multiprocessing systems; parallel algorithms; sparse matrices; antidiagonal wave pattern; approximate inverse computing; parallel algorithm; parallel normalized preconditioned conjugate gradient-type scheme; sparse finite element linear system; symmetric multiprocessor system; synchronization reduction; Concurrent computing; Distributed computing; Finite element methods; Gold; Linear systems; Multiprocessing systems; Parallel algorithms; Sparse matrices; Symmetric matrices; Vectors; OpenMP; parallel approximate inverses; parallel computations; parallel preconditioned conjugate gradient method; preconditioning; sparse linear systems; symmetric multiprocessor systems;

fLanguage

English

Publisher

ieee

Conference_Titel

Parallel and Distributed Computing, 2008. ISPDC '08. International Symposium on

Conference_Location

Krakow

Print_ISBN

978-0-7695-3472-5

Type

conf

DOI

10.1109/ISPDC.2008.18

Filename

4724252