Parallel Implementation in PC Clusters of a Lanczos-based Algorithm for an Electromagnetic Eigenvalue Problem

Author

Bernabeu, Miguel Oscar ; Garcia, Victor M. ; Taroncher, Máriam ; Vidal, Ana

Author_Institution

Dpt. de Sistemas Informaticos y Computacion, Univ. Politecnica de Valencia

fYear

2006

fDate

6-9 July 2006

Firstpage

296

Lastpage

300

Abstract

This paper describes a parallel implementation of a Lanczos-based method to solve generalised eigenvalue problems related to the modal computation of arbitrarily shaped waveguides. This efficient implementation is intended for execution in moderate-low cost workstations (2 to 4 processors). The problem under study has several features: the involved matrices are sparse with a certain structure, and all the eigenvalues needed are contained in a given interval. The novel parallel algorithms proposed show excellent speed-up for small number of processors

Keywords

eigenvalues and eigenfunctions; mathematics computing; message passing; parallel programming; sparse matrices; waveguide theory; workstation clusters; Lanczos-based algorithm; PC clusters; electromagnetic eigenvalue problem; parallel algorithms; sparse matrices; waveguides; Clustering algorithms; Concurrent computing; Costs; Eigenvalues and eigenfunctions; Electromagnetic scattering; Electromagnetic waveguides; Sparse matrices; Tellurium; Transmission line matrix methods; Workstations;

fLanguage

English

Publisher

ieee

Conference_Titel

Parallel and Distributed Computing, 2006. ISPDC '06. The Fifth International Symposium on

Conference_Location

Timisoara

Print_ISBN

0-7695-2638-1

Type

conf

DOI

10.1109/ISPDC.2006.39

Filename

4021940