The Mathematical Institute, University of Oxford, Eprints Archive

Fat vs. thin threading approach on GPUs: application to stochastic simulation of chemical reactions

Klingbeil, G. and Erban, R. and Giles, M. B. and Maini, P. K. (2010) Fat vs. thin threading approach on GPUs: application to stochastic simulation of chemical reactions. IEEE Transactions on Parallel & Distributed Systems . (Submitted)

[img]
Preview
PDF
466Kb

Abstract

We explore two different threading approaches on a graphics processing unit (GPU) exploiting two different characteristics of the current GPU architecture. The fat thread approach tries to minimise data access time by relying on shared memory and registers potentially sacrificing parallelism. The thin thread approach maximises parallelism and tries to hide access latencies. We apply these two approaches to the parallel stochastic simulation of chemical reaction systems using the stochastic simulation algorithm (SSA) by Gillespie (J. Phys. Chem, Vol. 81, p. 2340-2361, 1977). In these cases, the proposed thin thread approach shows comparable performance while eliminating the limitation of the reaction system’s size.

Item Type:Article
Subjects:D - G > General
Research Groups:Oxford Centre for Collaborative Applied Mathematics
ID Code:993
Deposited By:Peter Hudston
Deposited On:26 Oct 2010 10:23
Last Modified:09 Feb 2012 15:59

Repository Staff Only: item control page