Exact Sparse Matrix-Vector Multiplication on GPU's and Multicore Architectures

Brice Boyer 1 Jean-Guillaume Dumas 1 Pascal Giorgi 2
2 ARITH - Arithmétique informatique
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier
Abstract : We propose different implementations of the sparse matrix--dense vector multiplication (\spmv{}) for finite fields and rings $\Zb/m\Zb$. 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.
Complete list of metadatas

Cited literature [21 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-00475185
Contributor : Jean-Guillaume Dumas <>
Submitted on : Wednesday, April 21, 2010 - 3:15:32 PM
Last modification on : Thursday, July 4, 2019 - 9:54:02 AM
Long-term archiving on : Tuesday, September 28, 2010 - 12:28:10 PM

Files

ffspmv.pdf
Files produced by the author(s)

Identifiers

Citation

Brice Boyer, Jean-Guillaume Dumas, Pascal Giorgi. Exact Sparse Matrix-Vector Multiplication on GPU's and Multicore Architectures. PASCO'10: 4th International Symposium on Parallel Symbolic Computation, Jul 2010, Grenoble, France. pp.80-88, ⟨10.1145/1837210.1837224⟩. ⟨hal-00475185⟩

Share

Metrics

Record views

862

Files downloads

544