An implementation of a parallel iterative algorithm for the solution of large banded system on a cluster of workstations

M. Al-Towaiq, F. A M Masoud, A. B. Mnaouer, K. Day

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

In this paper, we present a parallel iterative solution for large banded systems of linear equations based on incomplete LU-factorization (ILU). A master-workers parallel computing scheme is used. The proposed algorithm incurs reduced storage and communication overhead as compared to previous methods. The reduction in communication overhead has been achieved by pipelining the interworkers message passing and limiting the communication between master and workers to initial work distribution and final results collection. The performance of the proposed algorithm is evaluated analytically and experimentally using Parallel Virtual Machine (PVM) as a parallel programming environment on a cluster of Linux workstations. The obtained performance results show that the proposed algorithm exhibits higher efficiency for coarse grain computations corresponding to large matrix sizes and reduced number of processors.

Original languageEnglish
Pages (from-to)378-386
Number of pages9
JournalInternational Journal of Modelling and Simulation
Volume28
Issue number4
Publication statusPublished - 2008

Fingerprint

Parallel Algorithms
Iterative Algorithm
Communication
Incomplete LU Factorization
Pipelining
Parallel programming
Programming Environments
Parallel Programming
Message passing
Virtual Machine
Iterative Solution
Linux
Parallel processing systems
Message Passing
Parallel Computing
System of Linear Equations
Linear equations
Factorization
High Efficiency
Limiting

Keywords

  • Banded systems
  • Cluster computing
  • Iterative methods
  • Parallel computing

ASJC Scopus subject areas

  • Hardware and Architecture
  • Software
  • Modelling and Simulation
  • Mechanics of Materials
  • Electrical and Electronic Engineering

Cite this

An implementation of a parallel iterative algorithm for the solution of large banded system on a cluster of workstations. / Al-Towaiq, M.; Masoud, F. A M; Mnaouer, A. B.; Day, K.

In: International Journal of Modelling and Simulation, Vol. 28, No. 4, 2008, p. 378-386.

Research output: Contribution to journalArticle

@article{9f64635c339b48c7bb8a00d3af01fc94,
title = "An implementation of a parallel iterative algorithm for the solution of large banded system on a cluster of workstations",
abstract = "In this paper, we present a parallel iterative solution for large banded systems of linear equations based on incomplete LU-factorization (ILU). A master-workers parallel computing scheme is used. The proposed algorithm incurs reduced storage and communication overhead as compared to previous methods. The reduction in communication overhead has been achieved by pipelining the interworkers message passing and limiting the communication between master and workers to initial work distribution and final results collection. The performance of the proposed algorithm is evaluated analytically and experimentally using Parallel Virtual Machine (PVM) as a parallel programming environment on a cluster of Linux workstations. The obtained performance results show that the proposed algorithm exhibits higher efficiency for coarse grain computations corresponding to large matrix sizes and reduced number of processors.",
keywords = "Banded systems, Cluster computing, Iterative methods, Parallel computing",
author = "M. Al-Towaiq and Masoud, {F. A M} and Mnaouer, {A. B.} and K. Day",
year = "2008",
language = "English",
volume = "28",
pages = "378--386",
journal = "International Journal of Modelling and Simulation",
issn = "0228-6203",
publisher = "ACTA Press",
number = "4",

}

TY - JOUR

T1 - An implementation of a parallel iterative algorithm for the solution of large banded system on a cluster of workstations

AU - Al-Towaiq, M.

AU - Masoud, F. A M

AU - Mnaouer, A. B.

AU - Day, K.

PY - 2008

Y1 - 2008

N2 - In this paper, we present a parallel iterative solution for large banded systems of linear equations based on incomplete LU-factorization (ILU). A master-workers parallel computing scheme is used. The proposed algorithm incurs reduced storage and communication overhead as compared to previous methods. The reduction in communication overhead has been achieved by pipelining the interworkers message passing and limiting the communication between master and workers to initial work distribution and final results collection. The performance of the proposed algorithm is evaluated analytically and experimentally using Parallel Virtual Machine (PVM) as a parallel programming environment on a cluster of Linux workstations. The obtained performance results show that the proposed algorithm exhibits higher efficiency for coarse grain computations corresponding to large matrix sizes and reduced number of processors.

AB - In this paper, we present a parallel iterative solution for large banded systems of linear equations based on incomplete LU-factorization (ILU). A master-workers parallel computing scheme is used. The proposed algorithm incurs reduced storage and communication overhead as compared to previous methods. The reduction in communication overhead has been achieved by pipelining the interworkers message passing and limiting the communication between master and workers to initial work distribution and final results collection. The performance of the proposed algorithm is evaluated analytically and experimentally using Parallel Virtual Machine (PVM) as a parallel programming environment on a cluster of Linux workstations. The obtained performance results show that the proposed algorithm exhibits higher efficiency for coarse grain computations corresponding to large matrix sizes and reduced number of processors.

KW - Banded systems

KW - Cluster computing

KW - Iterative methods

KW - Parallel computing

UR - http://www.scopus.com/inward/record.url?scp=58149250787&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=58149250787&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:58149250787

VL - 28

SP - 378

EP - 386

JO - International Journal of Modelling and Simulation

JF - International Journal of Modelling and Simulation

SN - 0228-6203

IS - 4

ER -