Faster, Cheaper, Better – a Hybridization Methodology to Develop Linear Algebra Software for GPUs

Emmanuel Agullo 1, 2 Cédric Augonnet 2, 3 Jack Dongarra 4 Hatem Ltaief 4 Raymond Namyst 2, 3 Samuel Thibault 2, 3 Stanimire Tomov 4
1 HiePACS - High-End Parallel Algorithms for Challenging Numerical Simulations
LaBRI - Laboratoire Bordelais de Recherche en Informatique, Inria Bordeaux - Sud-Ouest
3 RUNTIME - Efficient runtime systems for parallel architectures
Inria Bordeaux - Sud-Ouest, UB - Université de Bordeaux, CNRS - Centre National de la Recherche Scientifique : UMR5800
Abstract : In this chapter, we present a hybridization methodology for the development of linear algebra software for GPUs. The methodology is successfully used in MAGMA – a new generation of linear algebra libraries, similar in functionality to LAPACK, but extended for hybrid, GPU-based systems. Algorithms of interest are split into computational tasks. The tasks' execution is scheduled over the computational components of a hybrid system of multicore CPUs with GPU accelerators using StarPU – a runtime system for accelerator-based multicore architectures. StarPU enables to express parallelism through sequential-like code and schedules the different tasks over the hybrid processing units. The productivity becomes then fast and cheap as the development is high level, using existing software infrastructure. Moreover, the resulting hybrid algorithms are better performance-wise than corresponding homogeneous algorithms designed exclusively for either GPUs or homogeneous multicore CPUs.
Document type :
Book sections
Complete list of metadatas

Cited literature [9 references]  Display  Hide  Download

https://hal.inria.fr/inria-00547847
Contributor : Cédric Augonnet <>
Submitted on : Friday, December 17, 2010 - 3:06:36 PM
Last modification on : Thursday, December 20, 2018 - 3:36:07 PM
Long-term archiving on : Friday, March 18, 2011 - 2:47:54 AM

File

gpucomputinggems_plagma.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00547847, version 1

Citation

Emmanuel Agullo, Cédric Augonnet, Jack Dongarra, Hatem Ltaief, Raymond Namyst, et al.. Faster, Cheaper, Better – a Hybridization Methodology to Develop Linear Algebra Software for GPUs. Wen-mei W. Hwu. GPU Computing Gems, 2, Morgan Kaufmann, 2010. ⟨inria-00547847⟩

Share

Metrics

Record views

1241

Files downloads

483