HÉNON, Pascal; PELLEGRINI, Francois; RAMET, Pierre; ROMAN, Jean; SAAD, Yousef

Metadatos

Licencia de uso del documento

HÉNON, Pascal
Algorithms and high performance computing for grand challenge applications [SCALAPPLIX]

PELLEGRINI, Francois
Algorithms and high performance computing for grand challenge applications [SCALAPPLIX]
Laboratoire Bordelais de Recherche en Informatique [LaBRI]

RAMET, Pierre
Algorithms and high performance computing for grand challenge applications [SCALAPPLIX]
Laboratoire Bordelais de Recherche en Informatique [LaBRI]

Idioma

Communication dans un congrès

Este ítem está publicado en

Eleventh SIAM Conference on Parallel Processing for Scientific Computing, 2004, San Francisco. 2004

Resumen en inglés

PaStiX is a scientific library that provides a high performance direct supernodal solver for very large sparse linear systems. It relies on a block factorization based on an hybrid ordering (Nested Dissection + Halo Approximate Minimum Degree) obtained using the Scotch library. Efficient static scheduling and memory management are used to solve irregular problems with more of 25 millions unknowns on clusters of SMP nodes. In order to solve larger 3D problems, we apply these blockwise algorithms to compute robust and efficient parallel ILU preconditioners.< Leer menos

Metadatos

Compartir este ítem

Licencia de uso del documento

High Performance Complete and Incomplete Factorizations for Very Large Sparse Systems by using Scotch and PaStiX softwares

Idioma

Este ítem está publicado en

Resumen en inglés

URI

Orígen

Centros de investigación