On optimization techniques for the matrix multiplication on hybrid CPU+GPU platforms
-
Autores: Luis Pedro García, Javier Cuenca, Domingo Giménez Cánovas
- Localización: Annals of Multicore and GPU Programming: AMGP, ISSN 2341-3158, Vol. 1, Nº. 1, 2014, págs. 1-8
- Idioma: inglés
- Enlaces
-
Resumen
The use of auto-tuning techniques in a matrix multiplication routine for hybrid CPU+GPU platforms is analyzed. Basic models of the execution time of the hybrid routine and information obtained during its installation are used to optimize the execution time with a balanced assignation of the computation to the computing components in the heterogeneous system. Satisfactory results are obtained, with experimental execution times close to the lowest achievable
¿Es nuevo? Regístrese
Ventajas de registrarse
