Skip to Main content Skip to Navigation
Reports

Locality optimization on a NUMA architecture for hybrid LU factorization

Abstract : We study the impact of non-uniform memory accesses (NUMA) on the solution of dense general linear systems using an LU factorization algorithm. In particular we illustrate how an appropriate placement of the threads and memory on a NUMA architecture can improve the performance of the panel factorization and consequently accelerate the global LU factorization. We apply these placement strategies and present performance results for a hybrid multicore/GPU LU algorithm as it is implemented in the public domain library MAGMA.
Document type :
Reports
Complete list of metadata

Cited literature [23 references]  Display  Hide  Download

https://hal.inria.fr/hal-00957673
Contributor : Marc Baboulin <>
Submitted on : Monday, March 10, 2014 - 6:30:52 PM
Last modification on : Thursday, June 17, 2021 - 3:46:41 AM
Long-term archiving on: : Tuesday, June 10, 2014 - 12:50:42 PM

File

RR-8497.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00957673, version 1

Collections

Citation

Adrien Rémy, Marc Baboulin, Masha Sosonkina, Brigitte Rozoy. Locality optimization on a NUMA architecture for hybrid LU factorization. [Research Report] RR-8497, INRIA. 2014. ⟨hal-00957673⟩

Share

Metrics

Record views

927

Files downloads

409