Algebraic block red-black ordering method for parallelized ICCG solver with fast convergence and low communication costs

Author

Iwashita, Takeshi ; Shimasaki, Masaaki

Author_Institution

Acad. Center for Comput. & Media Studies, Kyoto Univ., Japan

Volume

39

Issue

3

fYear

2003

fDate

5/1/2003 12:00:00 AM

Firstpage

1713

Lastpage

1716

Abstract

Proposes a new parallelized incomplete Cholesky conjugate gradient (ICCG) solver effective on a small-scale multiprocessor system. The new method is based on a new reordering technique, namely the block red-black ordering method. Its parallel performance is evaluated in a finite edge-element eddy-current analysis. A numerical test shows that the proposed method is effective on a small number of processors due to fast convergence and low communication costs.

Keywords

conjugate gradient methods; convergence of numerical methods; eddy currents; electromagnetic fields; finite element analysis; EM field computations; algebraic block red-black ordering method; communication costs; convergence; electromagnetic field analyses; finite edge-element eddy-current analysis; incomplete Cholesky conjugate gradient solver; parallelized ICCG solver; reordering technique; small-scale multiprocessor system; Concurrent computing; Convergence; Costs; Finite difference methods; Kernel; Linear systems; Multiprocessing systems; Parallel processing; Performance analysis; Testing;

fLanguage

English

Journal_Title

Magnetics, IEEE Transactions on

Publisher

ieee

ISSN

0018-9464

Type

jour

DOI

10.1109/TMAG.2003.810531

Filename

1198563