Performance Analysis, Modeling and Prediction of a Parallel Multiblock Lattice Boltzmann Application Using Prophesy System Conference Paper uri icon

abstract

  • Recently, the Lattice Boltzmann method is widely used in simulating fluid flows. In this paper, we present the performance analysis, modeling and prediction of a parallel multiblock Lattice Boltzmann application on up to 512 processors on three SMP clusters: two IBM SP systems at San Diego Supercomputing Center (DataStar - p655 and p690) and one IBM SP system at the DOE National Energy Research Scientific Computing Center (Seaborg) using the Prophesy system. By characterizing the performance of the Lattice Boltzmann application as the problem size and the number of processors increase, we can identify and eliminate performance bottlenecks, and predict the application performance. The experimental results indicate that the application with large problem sizes scales well across these three clusters, and performance models using the coupling method are accurate with less than 4.8% average relative prediction error. © 2006 IEEE.

name of conference

  • 2006 IEEE International Conference on Cluster Computing

published proceedings

  • 2006 IEEE International Conference on Cluster Computing

author list (cited authors)

  • Wu, X., Taylor, V., Garrick, S., Yu, D., & Richard, J

citation count

  • 8

complete list of authors

  • Wu, Xingfu||Taylor, Valerie||Garrick, Shane||Yu, Dazhi||Richard, Jacques

publication date

  • September 2006

publisher