The application of parallel processing to the simplified marker and cell method

Author

Johnson, David B. ; Raad, Peter E. ; Rausch, Randall E.

Author_Institution

Dept. of Civil & Mech. Eng., Southern Methodist Univ., Dallas, TX, USA

fYear

1990

fDate

9-13 Dec 1990

Firstpage

357

Lastpage

362

Abstract

Significant improvements in the computational efficiency of the simulation of transient, free surface fluid flow problems by use of the simplified marker and cell method are described. The improvements are due to the replacement of the use of the successive overrelaxation method in a portion of the previous computer program with the use of the preconditioned conjugate gradient method and to the parallelization of major portions of the new program. The actual improvements in computational efficiency associated with the solution of five specific example problems are presented and discussed

Keywords

Navier-Stokes equations; civil engineering computing; computational complexity; conjugate gradient methods; flow simulation; mechanical engineering computing; parallel algorithms; physics computing; computational efficiency; conjugate gradient method; flow simulation; fluid flow; marker and cell method; parallel processing; parallelization; Algorithms; Computational efficiency; Computational modeling; Concurrent computing; Filling; Fluid flow; Navier-Stokes equations; Parallel processing; Surface treatment; Viscosity;

fLanguage

English

Publisher

ieee

Conference_Titel

Parallel and Distributed Processing, 1990. Proceedings of the Second IEEE Symposium on

Conference_Location

Dallas, TX

Print_ISBN

0-8186-2087-0

Type

conf

DOI

10.1109/SPDP.1990.143565

Filename

143565