Streaming Data from HDD to GPUs for Sustained Peak Performance
RWTH Aachen University, Aachen Institute for advanced study in Computational Engineering Science, Germany
arXiv:1302.4332 [cs.DC], (18 Feb 2013)
@article{2013arXiv1302.4332B,
author={Beyer}, L. and {Bientinesi}, P.},
title={"{Streaming Data from HDD to GPUs for Sustained Peak Performance}"},
journal={ArXiv e-prints},
archivePrefix={"arXiv"},
eprint={1302.4332},
primaryClass={"cs.DC"},
keywords={Computer Science – Distributed, Parallel, and Cluster Computing, Computer Science – Computational Engineering, Finance, and Science, Computer Science – Mathematical Software},
year={2013},
month={feb},
adsurl={http://adsabs.harvard.edu/abs/2013arXiv1302.4332B},
adsnote={Provided by the SAO/NASA Astrophysics Data System}
}
In the context of the genome-wide association studies (GWAS), one has to solve long sequences of generalized least-squares problems; such a task has two limiting factors: execution time often in the range of days or weeks and data management data sets in the order of Terabytes. We present an algorithm that obviates both issues. By pipelining the computation, and thanks to a sophisticated transfer strategy, we stream data from hard disk to main memory to GPUs and achieve sustained peak performance; with respect to a highly-optimized CPU implementation, our algorithm shows a speedup of 2.6x. Moreover, the approach lends itself to multiple GPUs and attains almost perfect scalability. When using 4 GPUs, we observe speedups of 9x over the aforementioned implementation, and 488x over a widespread biology library.
February 20, 2013 by hgpu
