Analysis and Performance of a Distributed Memory Multilevel Fast Multipole Algorithm

Author

Velamparambil, Sanjay ; Chew, Weng Cho

Author_Institution

Ansoft Corp., Boulder, CO, USA

Volume

53

Issue

8

fYear

2005

Firstpage

2719

Lastpage

2727

Abstract

In this paper, we analyze the communication pattern and study the scalability of a distributed memory implementation of the multilevel fast multipole algorithm (MLFMA) called $\\tt ScaleME$ . $\\tt ScaleME$ uses the message passing interface (MPI) for communication between processors. The parallelization of MLFMA uses a novel a hybrid scheme for distributing the workload across the processors. We study the communication and computational behavior and demonstrate the effectiveness of the parallelization scheme using realistic problems.

Keywords

application program interfaces; computational electromagnetics; distributed memory systems; electromagnetic wave scattering; integral equations; message passing; MLFMA parallelization; MPI; ScaleME; communication pattern; distributed memory; electromagnetic scattering; integral equation; message passing interface; multilevel fast multipole algorithm; Algorithm design and analysis; Antenna radiation patterns; Concurrent computing; Electromagnetic scattering; Integral equations; Large-scale systems; MLFMA; Message passing; Pattern analysis; Performance analysis; Electromagnetic scattering; fast multipole method (FMM); integral equations; parallel algorithms;

fLanguage

English

Journal_Title

Antennas and Propagation, IEEE Transactions on

Publisher

ieee

ISSN

0018-926X

Type

jour

DOI

10.1109/TAP.2005.851859

Filename

1492624