Landau Gauge Fixing on GPUs
CFTP, Departamento de Fisica, Instituto Superior Tecnico, Universidade Tecnica de Lisboa, Av. Rovisco Pais, 1049-001, Lisbon, Portugal
arXiv:1206.0675v1 [hep-lat] (4 Jun 2012)
@article{2012arXiv1206.0675C,
author={Cardoso}, N. and {Silva}, P.~J. and {Bicudo}, P. and {Oliveira}, O.},
title={"{Landau Gauge Fixing on GPUs}"},
journal={ArXiv e-prints},
archivePrefix={"arXiv"},
eprint={1206.0675},
primaryClass={"hep-lat"},
keywords={High Energy Physics – Lattice, Physics – Computational Physics},
year={2012},
month={jun},
adsurl={http://adsabs.harvard.edu/abs/2012arXiv1206.0675C},
adsnote={Provided by the SAO/NASA Astrophysics Data System}
}
In this paper we present and explore the performance of Landau gauge fixing in GPUs using CUDA. We consider the steepest descent algorithm with Fourier acceleration, and compare the GPU performance with a parallel CPU implementation. Using $32^4$ lattice volumes, we find that the computational power of a single Tesla C2070 GPU is equivalent to approximately 256 CPU cores.
June 5, 2012 by hgpu
