Performance of FORTRAN and C GPU Extensions for a Benchmark Suite of Fourier Pseudospectral Algorithms
Dept. of Physics, University of Michigan
arXiv:1206.3215v2 [physics.comp-ph] (12 Aug 2012)
@article{2012arXiv1206.3215C,
author={Cloutier}, B. and {Muite}, B.~K. and {Rigge}, P.},
title={"{A comparison of CPU and GPU performance for Fourier pseudospectral simulations of the Navier-Stokes, Cubic Nonlinear Schr$backslash$”${$o$}$dinger and Sine Gordon Equations}"},
journal={ArXiv e-prints},
archivePrefix={"arXiv"},
eprint={1206.3215},
primaryClass={"physics.comp-ph"},
keywords={Physics – Computational Physics, Computer Science – Performance},
year={2012},
month={jun},
adsurl={http://adsabs.harvard.edu/abs/2012arXiv1206.3215C},
adsnote={Provided by the SAO/NASA Astrophysics Data System}
}
A comparison of PGI OpenACC, FORTRAN CUDA, and Nvidia CUDA pseudospectral methods on a single GPU and GCC FORTRAN on single and multiple CPU cores is reported. The GPU implementations use CuFFT and the CPU implementations use FFTW. Porting pre-existing FORTRAN codes to utilize a GPUs is efficient and easy to implement with OpenACC and CUDA FORTRAN. Example programs are provided.
August 14, 2012 by hgpu
