Efficiency comparison of OpenMP, nVidia CUDA and StarPU technologies by the example of matrix multiplication

Authors

  • Konstantin Mikhaylovich Khankin South Ural State University

Keywords:

OpenMP, CUDA, StarPU technologies, energy saving

Abstract

In the article the description of OpenMP, nVidia CUDA and StarPU technologies, probable solutions of two matrix multiplication problem applying these technologies and the result of solution comparison by the criterion of resource consumption are considered.

Author Biography

Konstantin Mikhaylovich Khankin, South Ural State University

Post-Graduate Student of Electronic Computer Department of South Ural State University

References

OpenMP Application Program Interface. Version 3.1 July 2011. – http://www.openmp.org/mp-documents/OpenMP3.1.pdf

What is CUDA. – http://developer.nvidia.com/what-cuda

Tianhe-1A. – http://i.top500.org/system/176929

K computer. – http://i.top500.org/system/177232

StarPU. A Unified Runtime System for Heterogeneous Multicore Architectures. – http://runtime.bordeaux.inria.fr/StarPU/

Downloads

Published

2014-01-16

Issue

Vol. 13 No. 1 (2013)

Section

The main