Skip to Main content Skip to Navigation
Book sections

Towards a Unified CPU–GPU code hybridization: A GPU Based Optimization Strategy Efficient on Other Modern Architectures

Abstract : In this paper, we suggest a different methodology to shorten the code optimization development time while getting a unified code with good performance on different targeted devices. In the scope of this study, experiments are illustrated on a Discontinuous Galerkin code applied to Computational Fluid Dynamics. Tests are performed on CPUs, KNL Xeon-Phi and GPUs where performance comparison confirms that the GPU optimization guideline leads to efficient versions on CPU and Xeon-Phi for this kind of scientific applications. Based on these results, we finally suggest a methodology to end-up with an efficient hybridized CPU–GPU implementation.
Complete list of metadatas

Cited literature [9 references]  Display  Hide  Download

https://hal-centralesupelec.archives-ouvertes.fr/hal-01742774
Contributor : Stéphane Vialle <>
Submitted on : Monday, March 26, 2018 - 9:10:02 AM
Last modification on : Wednesday, September 16, 2020 - 5:48:14 PM
Long-term archiving on: : Thursday, September 13, 2018 - 8:33:24 AM

File

fullpaper.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01742774, version 1

Citation

Ludomir Oteski, Guillaume de Verdière, Sylvain Contassot-Vivier, Stephane Vialle, Juliet Ryan. Towards a Unified CPU–GPU code hybridization: A GPU Based Optimization Strategy Efficient on Other Modern Architectures. Parallel Computing is Everywhere,, 2018. ⟨hal-01742774⟩

Share

Metrics

Record views

543

Files downloads

245