Exact Sparse Matrix-Vector Multiplication on GPU’s and Multicore Architectures
Universite de Grenoble, Laboratoire Jean Kuntzmann, UMR CNRS 5224. 51, rue des Mathematiques, BP 53X, 38041 Grenoble, France
International Symposium on Parallel Symbolic Computation, Grenoble : France (2010); arXiv:1004.3719 [cs.DC] (21 Apr 2010)
@conference{boyer2010exact,
title={Exact sparse matrix-vector multiplication on GPU’s and multicore architectures},
author={Boyer, B. and Dumas, J.G. and Giorgi, P.},
booktitle={Proceedings of the 4th International Workshop on Parallel and Symbolic Computation},
pages={80–88},
year={2010},
organization={ACM}
}
We propose different implementations of the sparse matrix–dense vector multiplication (spmv{}) for finite fields and rings $Zb/mZb$. We take advantage of graphic card processors (GPU) and multi-core architectures. Our aim is to improve the speed of spmv{} in the linbox library, and henceforth the speed of its black box algorithms. Besides, we use this and a new parallelization of the sigma-basis algorithm in a parallel block Wiedemann rank implementation over finite fields.
November 11, 2010 by hgpu
