Load balancing for spatial-grid-based parallel numeric simulations on clusters of SMPs

Author

Gao, Huaien ; Schmidt, Andreas ; Gupta, Amitava ; Luksch, Peter

Author_Institution

Tech. Univ. Munich, Germany

fYear

2003

fDate

5-7 Feb. 2003

Firstpage

75

Lastpage

82

Abstract

Load distribution is an essential factor to parallel efficiency of numerical simulations that are based on spatial grids, especially on clusters of symmetric multiprocessors (SMP). This paper presents a method of mapping spatial grid nodes to processors that combines two load balancing methodologies, graph partitioning and graph matching, to achieve maximum parallel efficiency on SMP clusters. The method has been successfully applied to load distribution in a parallel computational fluid dynamics (CFD) simulation. Test runs on a PC cluster prove the effectiveness of our method.

Keywords

computational fluid dynamics; distributed memory systems; grid computing; numerical analysis; parallel programming; performance evaluation; resource allocation; CFD simulation; SMP clusters; computational fluid dynamics; graph matching; graph partitioning; load balancing; maximum parallel efficiency; numerical simulations; parallel numeric simulations; spatial grids; symmetric multiprocessors; Clustering algorithms; Computational fluid dynamics; Computational modeling; Costs; Grid computing; Load management; Mathematical programming; Numerical simulation; Switched-mode power supply; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Parallel, Distributed and Network-Based Processing, 2003. Proceedings. Eleventh Euromicro Conference on

Conference_Location

Genova, Italy

ISSN

1066-6192

Print_ISBN

0-7695-1875-3

Type

conf

DOI

10.1109/EMPDP.2003.1183569

Filename

1183569