Application of GPUs for the Calculation of Two Point Correlation Functions in Cosmology
Centro de Investigaciones Energeticas Medioambientales y Tecnologicas, Av.Complutense 40, 28040 Madrid, Spain
arXiv:1204.6630v1 [astro-ph.IM] (30 Apr 2012)
@article{2012arXiv1204.6630P,
author={Ponce}, R. and {Cardenas-Montes}, M. and {Rodriguez-Vazquez}, J.~J. and {Sanchez}, E. and {Sevilla}, I.},
title={"{Application of GPUs for the Calculation of Two Point Correlation Functions in Cosmology}"},
journal={ArXiv e-prints},
archivePrefix={"arXiv"},
eprint={1204.6630},
primaryClass={"astro-ph.IM"},
keywords={Astrophysics – Instrumentation and Methods for Astrophysics, Astrophysics – Cosmology and Extragalactic Astrophysics, Physics – Computational Physics},
year={2012},
month={apr},
adsurl={http://adsabs.harvard.edu/abs/2012arXiv1204.6630P},
adsnote={Provided by the SAO/NASA Astrophysics Data System}
}
In this work, we have explored the advantages and drawbacks of using GPUs instead of CPUs in the calculation of a standard 2-point correlation function algorithm, which is useful for the analysis of Large Scale Structure of galaxies. Taking into account the huge volume of data foreseen in upcoming surveys, our main goal has been to accelerate significantly the analysis codes. We find that GPUs offer a 100-fold increase in speed with respect to a single CPU without a significant deviation in the results. For comparison’s sake, an MPI version was developed as well. Some issues, like code implementation, which arise from using this option are discussed.
May 1, 2012 by hgpu
