Performance evaluation of a parallel lattice boltzmann method for cavity flows using cluster computing

Jun Ni, Ching Long Lin, Yongxiang Zhang, Tao He, Shaowen Wang, Boyd M. Knosp

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The Lattice Boltzmann Method (LBM) is a promising numerical approach in Computational Fluid Dynamics (CFD). LBM has relatively straightforward parallelism. Our research aims to analyze the performance of a parallel LBM algorithm for fluid flows on various LINUX clusters. This paper briefly depicts the algorithms for domain decomposition and data allocation in parallel LBM. Several data communication strategies using Message Passing Interface (MPI) library are discussed, and the effect of grid size on performance is analyzed. The paper also reports the benchmarks of parallel experiments conducted on different LINUX clusters (UI-ITS-32K and NCSA IA 32 LINUX clusters). The matching between the parallel results and the one using traditional Navier-Stokes approaches is demonstrated. In addition, this study provides basic insights into exploring LBM parallelism in a Grid-enhanced heterogeneous distributed environment.

Original languageEnglish (US)
Title of host publicationProceedings of the International Conference on Parallel and Distributed Processing Techniques and Applications, PDPTA'04
EditorsH.R. Arabnia
Pages10-16
Number of pages7
StatePublished - Dec 1 2004
EventProceedings of the International Conference on Parallel and Distributed Processing Techniques and Applications, PDPTA'04 - Las Vegas, NV, United States
Duration: Jun 21 2004Jun 24 2004

Publication series

NameProceedings of the International Conference on Parallel and Distributed Processing Techniques and Applications, PDPTA'04
Volume1

Other

OtherProceedings of the International Conference on Parallel and Distributed Processing Techniques and Applications, PDPTA'04
CountryUnited States
CityLas Vegas, NV
Period6/21/046/24/04

ASJC Scopus subject areas

  • Engineering(all)

Fingerprint Dive into the research topics of 'Performance evaluation of a parallel lattice boltzmann method for cavity flows using cluster computing'. Together they form a unique fingerprint.

  • Cite this

    Ni, J., Lin, C. L., Zhang, Y., He, T., Wang, S., & Knosp, B. M. (2004). Performance evaluation of a parallel lattice boltzmann method for cavity flows using cluster computing. In H. R. Arabnia (Ed.), Proceedings of the International Conference on Parallel and Distributed Processing Techniques and Applications, PDPTA'04 (pp. 10-16). (Proceedings of the International Conference on Parallel and Distributed Processing Techniques and Applications, PDPTA'04; Vol. 1).