2012
DOI: 10.1002/pamm.201210314
|View full text |Cite
|
Sign up to set email alerts
|

SPAI Preconditioners for HPC Applications

Abstract: Iterative methods to solve systems of linear equations Ax = b usually require preconditioners M to speed convergence. But the calculation of many preconditioners can be notoriously sequential. The sparse approximate inverse preconditioner (SPAI) has particular potential for high performance computing [1]. We have ported the SPAI algorithm to graphical processing units (GPUs) within NVIDIA's CUSP library [2] for sparse linear algebra. GPUs perform well on dense linear algebra problems where data resides for lon… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1

Citation Types

0
1
0

Year Published

2013
2013
2016
2016

Publication Types

Select...
2
1

Relationship

0
3

Authors

Journals

citations
Cited by 3 publications
(1 citation statement)
references
References 4 publications
0
1
0
Order By: Relevance
“…A substantial amount of research has been conducted on various preconditioning techniques for iterative solvers on GPUs including algebraic multigrid [Bell et al 2012;Gandham et al 2014;Richter et al 2014;Wagner et al 2012], incomplete factorizations [Li and Saad 2013;Naumov 2012], or sparse approximate inverses [Dehnavi et al 2013;Lukash et al 2012;Sawyer et al 2012]. Nevertheless, hardware-efficient and scalable black-box preconditioners for GPUs are not available, but instead the use of problem-specific information is required [Yokota et al 2011].…”
Section: Introductionmentioning
confidence: 99%
“…A substantial amount of research has been conducted on various preconditioning techniques for iterative solvers on GPUs including algebraic multigrid [Bell et al 2012;Gandham et al 2014;Richter et al 2014;Wagner et al 2012], incomplete factorizations [Li and Saad 2013;Naumov 2012], or sparse approximate inverses [Dehnavi et al 2013;Lukash et al 2012;Sawyer et al 2012]. Nevertheless, hardware-efficient and scalable black-box preconditioners for GPUs are not available, but instead the use of problem-specific information is required [Yokota et al 2011].…”
Section: Introductionmentioning
confidence: 99%