PUMA
Istituto di Matematica Applicata e Tecnologie Informatiche     
Fornasari N., Lanucara P., Rovida S. Parallelizzazione automatica di un codice GMRES a blocchi: Analisi delle prestazioni. Preprint ercim.cnr.ian//1998-1095, 1998.
 
 
Abstract
(English)
We discuss the performance of the automatic parallelization of a block GMRES algorithm, one of the most popular iterative solvers for multiple large non-symmetric linear systems. When the problem is large enough, the performance is quite satisfactory, due to the parallelization of the dense-to-dense matrix product. The detailed analysis of the code structure shows that another time consuming step of the algorithm is the dense-to-sparse matrix product. We test the explicit parallelization of this step by means of compiler directives, using GUIDE, a set of the standard OpenMP directives.
Subject GMRES a blocchi
Parallelizzazione automatica
G.1.0
G.1.3



Icona documento Open access Icona documento Restricted Icona documento Private

 


Per ulteriori informazioni, contattare: Librarian http://puma.isti.cnr.it

Valid HTML 4.0 Transitional