Preliminary work on graphics processing unit based direct simulation Monte Carlo

Authors:
Denis Gladkov;José-Juan Tapia;Roshan M. D'Souza
Affiliations:
College of Engineering & Applied Science, Milwaukee, WI;College of Engineering & Applied Science, Milwaukee, WI;College of Engineering & Applied Science, Milwaukee, WI
Venue:
Proceedings of the 2010 Conference on Grand Challenges in Modeling & Simulation
Year:
2010

Citing 6
Cited 1

Scalar and parallel optimized implementation of the direct simulation Monte Carlo method

Journal of Computational Physics
Mersenne twister: a 623-dimensionally equidistributed uniform pseudo-random number generator

ACM Transactions on Modeling and Computer Simulation (TOMACS) - Special issue on uniform random number generation
Parallel Programming in C with MPI and OpenMP

Parallel Programming in C with MPI and OpenMP
A comparison of CPUs, GPUs, FPGAs, and massively parallel processor arrays for random number generation

Proceedings of the ACM/SIGDA international symposium on Field programmable gate arrays
Designing efficient sorting algorithms for manycore GPUs

IPDPS '09 Proceedings of the 2009 IEEE International Symposium on Parallel&Distributed Processing
Programming Massively Parallel Processors: A Hands-on Approach

Programming Massively Parallel Processors: A Hands-on Approach

Graphics processing unit based direct simulation Monte Carlo

Simulation

Quantified Score

Hi-index	0.00

Visualization

Abstract

The Direct Simulation Monte Carlo (DSMC) method is used to simulate rarified fluid flow where the mean free path of molecules is of the same order of magnitude or greater than the representative physical length scale. The DSMC method models fluid flow by using particles which represent a large number of real molecules in a probabilistic simulation to solve the Boltzmann equation. Due to the large number of individual particles that must be represented, the computational complexity of the DSMC method is beyond the capabilities of desktop computing. In this paper we present data-parallel techniques to efficiently simulate large-scale DSMC models on graphics processing units (GPUs). We have designed and implemented the Nanbu DSMC method to simulate gas flow. Our results show a substantial gain in performance compared to the serial version running on a single CPU.