LU factorization for accelerator-based systems

Emmanuel Agullo, Cédric Augonnet, Jack Dongarra, Mathieu Faverge, Julien Langou, Hatem Ltaief, Stanimire Tomov

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

38 Scopus citations

Abstract

Multicore architectures enhanced with multiple GPUs are likely to become mainstream High Performance Computing (HPC) platforms in a near future. In this paper, we present the design and implementation of an LU factorization using tile algorithm that can fully exploit the potential of such platforms in spite of their complexity. We use a methodology derived from previous work on Cholesky and QR factorizations. Our contributions essentially consist of providing new CPU/GPU hybrid LU kernels, studying the impact on performance of the looking variants as well as the storage layout in presence of pivoting, tuning the kernels for two different machines composed of multiple recent NVIDIA Tesla S1070 (four GPUs total) and Fermi-based S2050 GPUs (three GPUs total), respectively. The hybrid tile LU asymptotically achieves 1 Tflop/s in single precision on both hardwares. The performance in double precision arithmetic reaches 500 Gflop/s on the Fermi-based system, twice faster than the old GPU generation of Tesla S1070. We also discuss the impact of the number of tiles on the numerical stability. We show that the numerical results of the tile LU factorization will be accurate enough for most applications as long as the computations are performed in double precision arithmetic.

Original languageEnglish
Title of host publicationProceedings of the 2011 9th IEEE/ACS International Conference on Computer Systems and Applications, AICCSA 2011
Pages217-224
Number of pages8
DOIs
StatePublished - 2011
Externally publishedYes
Event9th ACS/IEEE International Conference on Computer Systems and Applications, AICCSA 2011 - Sharm El-Sheikh, Egypt
Duration: Dec 27 2011Dec 30 2011

Publication series

NameProceedings of IEEE/ACS International Conference on Computer Systems and Applications, AICCSA
ISSN (Print)2161-5322
ISSN (Electronic)2161-5330

Conference

Conference9th ACS/IEEE International Conference on Computer Systems and Applications, AICCSA 2011
Country/TerritoryEgypt
CitySharm El-Sheikh
Period12/27/1112/30/11

Keywords

  • Dense Linear Algebra
  • High Performance Computing
  • Hybrid Architecture
  • LU Factorization
  • Multicore
  • Multiple GPU Accelerators
  • Numerical Accuracy
  • Tile Algorithm

Fingerprint

Dive into the research topics of 'LU factorization for accelerator-based systems'. Together they form a unique fingerprint.

Cite this