Abstract
New results concerning a recently introduced parallel preconditioner for the solution of large nonconforming FE linear systems are presented. The studied algorithm is based on the modified incomplete Cholesky factorization MIC(0) applied to a locally constructed approximation of the original stiffness matrix. The overlapping of communications and computations is possible due to a suitable reordering of the computations applied in the MPI code. The obtained improvement of the real performance is illustrated by numerical tests on a Beowulf-type Linux cluster and on a Sun Fire symmetric multiprocessor. © Springer-Verlag Berlin Heidelberg 2006.
Cite
CITATION STYLE
Bencheva, G., Margenov, S., & Starý, J. (2006). Parallel PCG solver for nonconforming FE problems: Overlapping of communications and computations. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 3743 LNCS, pp. 646–654). https://doi.org/10.1007/11666806_74
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
