Scalability analysis of partitioning strategies for finite element graphs: a summary of results

Author

Grama, Ananth Y. ; Kumar, Vipin

Author_Institution

Dept. of Comput. Sci., Minnesota Univ., Minneapolis, MN, USA

fYear

1992

fDate

16-20 Nov 1992

Firstpage

83

Lastpage

92

Abstract

The authors present a scalability analysis of three partitioning strategies, namely striped partitioning, binary decomposition, and scattered decomposition. The analysis is performed using the Isoefficiency metric, which helps in predicting the performances of these schemes on a range of processors and architectures. The performance of each of these schemes is related to the various problem characteristics such as mesh geometry and density. Isoefficiencies are presented for hypercube and mesh connected architectures. Theoretical results are verified through simulations

Keywords

finite element analysis; graph theory; mesh generation; Isoefficiency metric; binary decomposition; finite element graphs; hypercube; mesh connected architectures; mesh geometry; partitioning strategies; scalability analysis; scattered decomposition; striped partitioning; Availability; Computational modeling; Computer science; Differential equations; Finite element methods; High performance computing; Large-scale systems; Partitioning algorithms; Performance analysis; Scalability;

fLanguage

English

Publisher

ieee

Conference_Titel

Supercomputing '92., Proceedings

Conference_Location

Minneapolis, MN

Print_ISBN

0-8186-2630-5

Type

conf

DOI

10.1109/SUPERC.1992.236707

Filename

236707