Parallelisation of DEM-LBM using domain decomposition

Jun Li; Xilin Cui; Andrew Chan; John Bridgeman

doi:10.4028/www.scientific.net/AMM.553.531

Parallelisation of DEM-LBM using domain decomposition

Jun Li^*, Xilin Cui, Andrew Chan, John Bridgeman

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

The coupled Discrete Element (DEM) - Lattice Boltzmann Method (LBM) often suffers from the high computational cost due to the fine meshes used. The intrinsic parallel nature of DEM-LBM makes it possible to be applied to the analysis of realistic engineering problems using parallel computer. In this paper the domain decomposition is implemented in the validated coupled DEM-LBM code FPS-BHAM and its performance is tested. A parallel efficiency of 0.72 has been achieved by using 32 processors, which shows a very good parallel behaviour of the DEM-LBM. Besides, the feature of 'pseudo-vector processing capacity' boosts the parallel behaviour of LBM with domain decomposition.

Original language	English
Title of host publication	Advances in Computational Mechanics
Publisher	Trans Tech Publications Inc
Pages	531-536
Number of pages	6
Volume	553
ISBN (Print)	9783038350682
DOIs	https://doi.org/10.4028/www.scientific.net/AMM.553.531
Publication status	Published - 2014
Event	1st Australasian Conference on Computational Mechanics, ACCM 2013 - Sydney, NSW, Australia Duration: 3 Oct 2013 → 4 Oct 2013

Publication series

Name	Applied Mechanics and Materials
Volume	553
ISSN (Print)	16609336
ISSN (Electronic)	16627482

Conference

Conference	1st Australasian Conference on Computational Mechanics, ACCM 2013
Country/Territory	Australia
City	Sydney, NSW
Period	3/10/13 → 4/10/13

Keywords

Discrete element method
Domain decomposition
Lattice boltazmann method
Parallelisation

ASJC Scopus subject areas

General Engineering

Access to Document

10.4028/www.scientific.net/AMM.553.531

Cite this

@inproceedings{6a6e42e07e234cb9a73617000a5553da,

title = "Parallelisation of DEM-LBM using domain decomposition",

abstract = "The coupled Discrete Element (DEM) - Lattice Boltzmann Method (LBM) often suffers from the high computational cost due to the fine meshes used. The intrinsic parallel nature of DEM-LBM makes it possible to be applied to the analysis of realistic engineering problems using parallel computer. In this paper the domain decomposition is implemented in the validated coupled DEM-LBM code FPS-BHAM and its performance is tested. A parallel efficiency of 0.72 has been achieved by using 32 processors, which shows a very good parallel behaviour of the DEM-LBM. Besides, the feature of 'pseudo-vector processing capacity' boosts the parallel behaviour of LBM with domain decomposition.",

keywords = "Discrete element method, Domain decomposition, Lattice boltazmann method, Parallelisation",

author = "Jun Li and Xilin Cui and Andrew Chan and John Bridgeman",

year = "2014",

doi = "10.4028/www.scientific.net/AMM.553.531",

language = "English",

isbn = "9783038350682",

volume = "553",

series = "Applied Mechanics and Materials",

publisher = "Trans Tech Publications Inc",

pages = "531--536",

booktitle = "Advances in Computational Mechanics",

address = "Germany",

note = "1st Australasian Conference on Computational Mechanics, ACCM 2013 ; Conference date: 03-10-2013 Through 04-10-2013",

}

Li, J, Cui, X, Chan, A & Bridgeman, J 2014, Parallelisation of DEM-LBM using domain decomposition. in Advances in Computational Mechanics. vol. 553, Applied Mechanics and Materials, vol. 553, Trans Tech Publications Inc, pp. 531-536, 1st Australasian Conference on Computational Mechanics, ACCM 2013, Sydney, NSW, Australia, 3/10/13. https://doi.org/10.4028/www.scientific.net/AMM.553.531

TY - GEN

T1 - Parallelisation of DEM-LBM using domain decomposition

AU - Li, Jun

AU - Cui, Xilin

AU - Chan, Andrew

AU - Bridgeman, John

PY - 2014

Y1 - 2014

N2 - The coupled Discrete Element (DEM) - Lattice Boltzmann Method (LBM) often suffers from the high computational cost due to the fine meshes used. The intrinsic parallel nature of DEM-LBM makes it possible to be applied to the analysis of realistic engineering problems using parallel computer. In this paper the domain decomposition is implemented in the validated coupled DEM-LBM code FPS-BHAM and its performance is tested. A parallel efficiency of 0.72 has been achieved by using 32 processors, which shows a very good parallel behaviour of the DEM-LBM. Besides, the feature of 'pseudo-vector processing capacity' boosts the parallel behaviour of LBM with domain decomposition.

AB - The coupled Discrete Element (DEM) - Lattice Boltzmann Method (LBM) often suffers from the high computational cost due to the fine meshes used. The intrinsic parallel nature of DEM-LBM makes it possible to be applied to the analysis of realistic engineering problems using parallel computer. In this paper the domain decomposition is implemented in the validated coupled DEM-LBM code FPS-BHAM and its performance is tested. A parallel efficiency of 0.72 has been achieved by using 32 processors, which shows a very good parallel behaviour of the DEM-LBM. Besides, the feature of 'pseudo-vector processing capacity' boosts the parallel behaviour of LBM with domain decomposition.

KW - Discrete element method

KW - Domain decomposition

KW - Lattice boltazmann method

KW - Parallelisation

UR - http://www.scopus.com/inward/record.url?scp=84902074111&partnerID=8YFLogxK

U2 - 10.4028/www.scientific.net/AMM.553.531

DO - 10.4028/www.scientific.net/AMM.553.531

M3 - Conference contribution

AN - SCOPUS:84902074111

SN - 9783038350682

VL - 553

T3 - Applied Mechanics and Materials

SP - 531

EP - 536

BT - Advances in Computational Mechanics

PB - Trans Tech Publications Inc

T2 - 1st Australasian Conference on Computational Mechanics, ACCM 2013

Y2 - 3 October 2013 through 4 October 2013

ER -

Parallelisation of DEM-LBM using domain decomposition

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this