Particle method on GPU
Laboratoire Jean Kuntzmann (LJK), Universite Joseph Fourier – Grenoble I – Universite Pierre Mendes-France – Grenoble II – Institut Polytechnique de Grenoble – Grenoble Institute of Technology
hal-00780565, 24 January 2013
@unpublished{etancelin:hal-00780565,
hal_id={hal-00780565},
url={http://hal.archives-ouvertes.fr/hal-00780565},
title={Particle method on GPU},
author={Etancelin, Jean-Mathieu and Cottet, Georges-Henri and Picard, Christophe and P{‘e}rignon, Franck},
keywords={particle method, advection equation, GPU, opencl, python},
language={Anglais},
affiliation={Laboratoire Jean Kuntzmann – LJK},
note={CANUM 2012. To appear in ESAIM Proceedings},
year={2013},
month={Sep},
pdf={http://hal.archives-ouvertes.fr/hal-00780565/PDF/ms_PM_GPU.pdf}
}
In this article we present a graphics processing unit (GPU) implementation of a particle method for transport equations. More precisely the numerical method under consideration is a remeshed particle method. Not only remeshing particles makes simulations more accurate in flows with strong strain, but it leads to algorithms more regular in term of data structures. In this work, we develop a Python library using GPU through OpenCL standard that implements this remeshed particle method which already shows interesting performances.
January 31, 2013 by hgpu
