A simple OpenMP scheme for parallel iteration solvers in finite element analysis

Research output: Contribution to journalArticle

13 Citations (Scopus)

Abstract

This study develops an OpenMP scheme to parallel the preconditioned conjugate gradient methods (PCG) in shared memory computers. The proposed method is simple and systematic, so a minor change in traditional PCG methods may produce effective parallelism. At first, the global stiffness matrix is renumbered in order to produce a parallel three-line form matrix, and a subroutine only needs to be called once in the finite element analysis. Several basic OpenMP commands are then added into the traditional incomplete Cholesky factorization (ILU) and symmetric successive over-relaxation (SSOR) codes to make the procedures of matrix multiplication, decomposition, forward substitution, and backward substitution fully parallel.

Original languageEnglish
Pages (from-to)91-108
Number of pages18
JournalCMES - Computer Modeling in Engineering and Sciences
Volume64
Issue number1
Publication statusPublished - 2010 Oct 11

Fingerprint

Conjugate gradient method
OpenMP
Preconditioned Conjugate Gradient Method
Substitution reactions
Finite Element
Iteration
Finite element method
Substitution
Subroutines
Stiffness matrix
Factorization
Cholesky factorisation
Matrix multiplication
Stiffness Matrix
Shared Memory
Decomposition
Data storage equipment
Parallelism
Minor
Decompose

All Science Journal Classification (ASJC) codes

  • Software
  • Modelling and Simulation
  • Computer Science Applications

Cite this

@article{1ed28f6c941540b1a3a99041117aa2c6,
title = "A simple OpenMP scheme for parallel iteration solvers in finite element analysis",
abstract = "This study develops an OpenMP scheme to parallel the preconditioned conjugate gradient methods (PCG) in shared memory computers. The proposed method is simple and systematic, so a minor change in traditional PCG methods may produce effective parallelism. At first, the global stiffness matrix is renumbered in order to produce a parallel three-line form matrix, and a subroutine only needs to be called once in the finite element analysis. Several basic OpenMP commands are then added into the traditional incomplete Cholesky factorization (ILU) and symmetric successive over-relaxation (SSOR) codes to make the procedures of matrix multiplication, decomposition, forward substitution, and backward substitution fully parallel.",
author = "Ju, {S. H.}",
year = "2010",
month = "10",
day = "11",
language = "English",
volume = "64",
pages = "91--108",
journal = "CMES - Computer Modeling in Engineering and Sciences",
issn = "1526-1492",
publisher = "Tech Science Press",
number = "1",

}

A simple OpenMP scheme for parallel iteration solvers in finite element analysis. / Ju, S. H.

In: CMES - Computer Modeling in Engineering and Sciences, Vol. 64, No. 1, 11.10.2010, p. 91-108.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A simple OpenMP scheme for parallel iteration solvers in finite element analysis

AU - Ju, S. H.

PY - 2010/10/11

Y1 - 2010/10/11

N2 - This study develops an OpenMP scheme to parallel the preconditioned conjugate gradient methods (PCG) in shared memory computers. The proposed method is simple and systematic, so a minor change in traditional PCG methods may produce effective parallelism. At first, the global stiffness matrix is renumbered in order to produce a parallel three-line form matrix, and a subroutine only needs to be called once in the finite element analysis. Several basic OpenMP commands are then added into the traditional incomplete Cholesky factorization (ILU) and symmetric successive over-relaxation (SSOR) codes to make the procedures of matrix multiplication, decomposition, forward substitution, and backward substitution fully parallel.

AB - This study develops an OpenMP scheme to parallel the preconditioned conjugate gradient methods (PCG) in shared memory computers. The proposed method is simple and systematic, so a minor change in traditional PCG methods may produce effective parallelism. At first, the global stiffness matrix is renumbered in order to produce a parallel three-line form matrix, and a subroutine only needs to be called once in the finite element analysis. Several basic OpenMP commands are then added into the traditional incomplete Cholesky factorization (ILU) and symmetric successive over-relaxation (SSOR) codes to make the procedures of matrix multiplication, decomposition, forward substitution, and backward substitution fully parallel.

UR - http://www.scopus.com/inward/record.url?scp=77957588977&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77957588977&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:77957588977

VL - 64

SP - 91

EP - 108

JO - CMES - Computer Modeling in Engineering and Sciences

JF - CMES - Computer Modeling in Engineering and Sciences

SN - 1526-1492

IS - 1

ER -