A Simple Guideline for Code Optimizations on Modern Architectures with OpenACC and CUDA

Ludomir Oteski 1 Guillaume Colin-De-Verdière 2 Sylvain Contassot-Vivier 3, 4 Stephane Vialle 5, 6, 7, 8 Juliet Ryan 1
1 Palaiseau - ONERA - The French Aerospace Lab
2 DAM/DIF - DAM Île-de-France
3 UL - Université de Lorraine
4 LORIA - Laboratoire Lorrain de Recherche en Informatique et ses Applications
5 Université Paris-Saclay
6 CentraleSupélec
7 UMI2958 - Georgia Tech - CNRS [Metz]
8 ParSys - LRI - Systèmes parallèles (LRI)
LRI - Laboratoire de Recherche en Informatique
Abstract : Learn a simple strategy guideline to optimize applications runtime. The strategy is based on four steps and illustrated on a two-dimensional Discontinuous Galerkin solver for computational fluid dynamics on structured meshes. Starting from a CPU sequential code, we guide the audience through the different steps that allowed us to increase performances on a GPU around 149 times the original runtime of the code (performances evaluated on a K20Xm). The same optimization strategy is applied to the CPU code and increases performances around 35 times the original run time (performances evaluated on a E5-1650v3 processor). Finally, different hardware architectures (Xeon CPUs, GPUs, KNL) are benchmarked with the native CUDA implementation and one based on OpenACC.
Keywords : CUDA guideline OpenACC Code optimization GPU Xeon-phi
Type de document :
Communication dans un congrès
GPU Technology Conference 2017, May 2017, San Jose, United States
Liste complète des métadonnées
https://hal-centralesupelec.archives-ouvertes.fr/hal-01527684
Contributeur : Stéphane Vialle <>
Soumis le : mardi 30 mai 2017 - 11:52:14
Dernière modification le : lundi 15 octobre 2018 - 15:54:03

Annexe

Identifiants

  • HAL Id : hal-01527684, version 1

Collections

CENTRALESUPELEC | UNIV-FCOMTE | UMR8623 | INRIA | UMI-GTL | CEA | LORIA | UNIV-LORRAINE | UNIV-PARIS-SACLAY | INRIA2017 | ONERA | IDMAD | UMI-COMPUTERSCIENCE | LRI-PARSYS | CEA-UPSAY | DAM

Citation

Ludomir Oteski, Guillaume Colin-De-Verdière, Sylvain Contassot-Vivier, Stephane Vialle, Juliet Ryan. A Simple Guideline for Code Optimizations on Modern Architectures with OpenACC and CUDA. GPU Technology Conference 2017, May 2017, San Jose, United States. 〈hal-01527684〉

Exporter

BibTeX TEI DC DCterms EndNote

Partager

Métriques

Consultations de la notice

503

Téléchargements de fichiers

73

Contact

support.ccsd.cnrs.fr
hal.support@ccsd.cnrs.fr
Logo CCSD