High performance grid and cluster computing for some optimization problems

Author

Fujisawa, Katsuki ; Kojima, Masakazu ; Takeda, Akiko ; Yamashita, Makoto

Author_Institution

Dept. of Math. Sci., Tokyo Denki Univ., Japan

fYear

2004

fDate

26-30 Jan. 2004

Firstpage

612

Lastpage

615

Abstract

The aim of this short article is to show that grid and cluster computing provides tremendous power to optimization methods. The methods that the article picks up are a successive convex relaxation method for quadratic optimization problems, a polyhedral homotopy method for polynomial systems of equations and a primal-dual interior-point method for semidefinite programming problems. Their parallel implementations on grids and clusters together with numerical results are reported.

Keywords

grid computing; optimisation; workstation clusters; high performance cluster computing; high performance grid computing; optimization methods; parallel implementations; polyhedral homotopy method; polynomial systems; primal-dual interior-point method; quadratic optimization problems; semidefinite programming problems; successive convex relaxation method; Computer networks; Distributed computing; Equations; Grid computing; High performance computing; Large-scale systems; Optimization methods; Polynomials; Power generation economics; Relaxation methods;

fLanguage

English

Publisher

ieee

Conference_Titel

Applications and the Internet Workshops, 2004. SAINT 2004 Workshops. 2004 International Symposium on

Print_ISBN

0-7695-2050-2

Type

conf

DOI

10.1109/SAINTW.2004.1268696

Filename

1268696