Chapitre D'ouvrage Année : 2018

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

Résumé

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.

Fichier principal
Vignette du fichier
fullpaper.pdf (789.96 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence
Loading...

Dates et versions

hal-01742774 , version 1 (26-03-2018)

Licence

Identifiants

Citer

Ludomir Oteski, Guillaume Colin 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. Sanzio Bassini, Marco Danelutto, Patrizio Dazzi, Gerhard R. Joubert, Frans Peters. Parallel Computing is Everywhere, 32, pp.390-399, 2018, 978-1-61499-842-6. ⟨10.3233/978-1-61499-843-3-390⟩. ⟨hal-01742774⟩
1580 Consultations
840 Téléchargements

Altmetric

Partager

  • More