Dynamically scheduled Cholesky factorization on multicore architectures with GPU accelerators.

Abstract : Although the hardware has dramatically changed in the last few years, nodes of multicore chips augmented by Graphics Processing Units (GPUs) seem to be a trend of major importance. Previous approaches for scheduling dense linear operations on such a complex node led to high performance but at the double cost of not using the potential of all the cores and producing a static and non generic code. In this extended abstract, we present a new approach for scheduling dense linear algebra operations on multicore architectures with GPU accelerators using a dynamic scheduler capable of using the full potential of the node [1]. We underline the benefits both in terms of programmability and performance. We illustrate our approach with a Cholesky factorization relying on cutting edge GPU and CPU kernels [2], [3] achieving roughly 900 Gflop/s on an eight cores node accelerated with three NVIDIA Tesla GPUs.
Complete list of metadatas

Cited literature [3 references]  Display  Hide  Download

https://hal.inria.fr/inria-00547616
Contributor : Samuel Thibault <>
Submitted on : Thursday, December 16, 2010 - 7:09:19 PM
Last modification on : Thursday, August 22, 2019 - 2:42:03 PM
Long-term archiving on : Monday, November 5, 2012 - 2:30:26 PM

File

saahpc.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00547616, version 1

Citation

Emmanuel Agullo, Cédric Augonnet, Jack Dongarra, Hatem Ltaief, Raymond Namyst, et al.. Dynamically scheduled Cholesky factorization on multicore architectures with GPU accelerators.. Symposium on Application Accelerators in High Performance Computing (SAAHPC), Jul 2010, Knoxville, United States. ⟨inria-00547616⟩

Share

Metrics

Record views

834

Files downloads

577