A parallel algorithm to solve large stiff ODE systems on grid systems

Author

Bahi, Jacques ; Charr, Jean-Claude ; Couturier, Raphaël ; Laiymani, David

Author_Institution

Lab. d´´Inf. de l´´Univ. de Franche-Comte, Univ. of Franche-Comte, Belfort

fYear

2007

fDate

17-20 Sept. 2007

Firstpage

534

Lastpage

541

Abstract

This paper introduces a parallel algorithm to solve large stiff ODE systems in a geographically distant cluster environment. This algorithm is based on the coupling of the waveform relaxation concept and the CVODE algorithm. With respect to the standard PVODE algorithm, it allows to drastically reduce the number of messages exchanged between nodes. It is a coarse grained algorithm well suited for distant grid environments connected via high latency networks. In this paper our work consists in analyzing the execution times taken by the PVODE solver and our algorithm and in explaining the benefits brought by this work.

Keywords

grid computing; parallel algorithms; coarse grained algorithm; geographically distant cluster environment; grid systems; high latency networks; large stiff ODE systems; parallel algorithm; waveform relaxation concept; Algorithm design and analysis; Clustering algorithms; Computer architecture; Delay; Differential equations; Distributed computing; Hardware; Parallel algorithms; Power system reliability; Supercomputers;

fLanguage

English

Publisher

ieee

Conference_Titel

Cluster Computing, 2007 IEEE International Conference on

Conference_Location

Austin, TX

ISSN

1552-5244

Print_ISBN

978-1-4244-1387-4

Electronic_ISBN

1552-5244

Type

conf

DOI

10.1109/CLUSTR.2007.4629280

Filename

4629280