Códigos Numéricos para el Producto de Matrices Ralas

Horacio E. Retamales, Carlos H. Delahaye, Luis E. Sánchez

Abstract


A set of subroutines have been developed, in order to min1mlze the CPO time and storinq requirements in processinq block matrix products using non zero elements of each block matrix or non zero rows (columns).
Comparation is set among product of blocks considered as sparse matrixes (column matrixes) or complete matrixes.
Advantajes are ahown by results been obtained, in using developed techniques.

Full Text:

PDF



Asociación Argentina de Mecánica Computacional
Güemes 3450
S3000GLN Santa Fe, Argentina
Phone: 54-342-4511594 / 4511595 Int. 1006
Fax: 54-342-4511169
E-mail: amca(at)santafe-conicet.gov.ar
ISSN 2591-3522