Parallel direct simulation Monte Carlo computation using CUDA on GPUs

C. C. Su, C. W. Hsieh, M. R. Smith, M. C. Jermy, J. S. Wu

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

10 Citations (Scopus)

Abstract

In this study computations of the two-dimensional Direct Simulation Monte Carlo (DSMC) method using Graphics Processing Units (GPUs) are presented. An all-device (GPU) computational approach is adopted-where the entire computation is performed on the GPU device, leaving the CPU idle-which includes particle moving, indexing, collisions between particles and state sampling. The subsequent application to GPU computation requires various changes to the original DSMC method to ensure efficient performance on the GPU device. Communications between the host (CPU) and device (GPU) occur only during problem initialization and simulation conclusion when results are only copied from the device to the host. Several multi-dimensional benchmark tests are employed to demonstrate the correctness of the DSMC implementation. We demonstrate here the application of DSMC using a single-GPU, with speedups of 3∼10 times as compared to a high-end Intel CPU (Intel Xeon X5472) depending upon the size and the level of rarefaction encountered in the simulation.

Original languageEnglish
Title of host publication27th International Symposium on Rarefied Gas Dynamics - 2010, RGD27
Pages343-347
Number of pages5
EditionPART 1
DOIs
Publication statusPublished - 2011
Event27th International Symposium on Rarefied Gas Dynamics, RGD27 - Pacific Grove, CA, United States
Duration: 2011 Jul 102011 Jul 15

Publication series

NameAIP Conference Proceedings
NumberPART 1
Volume1333
ISSN (Print)0094-243X
ISSN (Electronic)1551-7616

Other

Other27th International Symposium on Rarefied Gas Dynamics, RGD27
Country/TerritoryUnited States
CityPacific Grove, CA
Period11-07-1011-07-15

All Science Journal Classification (ASJC) codes

  • Physics and Astronomy(all)

Fingerprint

Dive into the research topics of 'Parallel direct simulation Monte Carlo computation using CUDA on GPUs'. Together they form a unique fingerprint.

Cite this