Parallel out-of-core matrix inversion

Abstract : This paper presents a parallel out-of-core algorithm to invert huge matrices, that is when size of matrices is larger than the available physical memory by one or more orders of magnitude. Preliminary performance results are shown for a commodity cluster. An accurate prediction performance model of the algorithm is given. Thanks to the prediction model, optimizations which avoid the overhead of the out-of-core algorithm are derived. Performances of the optimized algorithm using a O(N) memory size are similar to the performances of the best known parallel in-core algorithm using a O(N^2) memory size (where N is the matrix order). There is no memory restriction for inversion of huge matrices.
Document type :
Reports
Complete list of metadatas

https://hal-lara.archives-ouvertes.fr/hal-02101901
Contributor : Colette Orange <>
Submitted on : Wednesday, April 17, 2019 - 9:09:20 AM
Last modification on : Wednesday, November 20, 2019 - 7:24:05 AM

File

RR2002-04.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02101901, version 1

Collections

Citation

Eddy Caron, Gil Utard. Parallel out-of-core matrix inversion. [Research Report] LIP RR-2002-04, Laboratoire de l'informatique du parallélisme. 2002, 2+17p. ⟨hal-02101901⟩

Share

Metrics

Record views

11

Files downloads

32