Towards seismic wave modeling on heterogeneous many-core architectures using task-based runtime system

Víctor Martínez 1 David Michéa 2 Fabrice Dupros 2 Olivier Aumage 3 Samuel Thibault 4, 3 Hideo Aochi 2 Philippe Olivier Alexandre Navaux 1
1 UFRGS - Instituto de Informática da UFRGS
2 BRGM - Bureau de Recherches Géologiques et Minières (BRGM)
3 STORM - STatic Optimizations, Runtime Methods
LaBRI - Laboratoire Bordelais de Recherche en Informatique, Inria Bordeaux - Sud-Ouest
4 LaBRI - Laboratoire Bordelais de Recherche en Informatique
Abstract : Understanding three-dimensional seismic wave propagation in complex media remains one of the main challenges of quantitative seismology. Because of its simplicity and numerical efficiency, the finite-differences method is one of the standard techniques implemented to consider the elastodynamics equation. Additionally, this class of modelling heavily relies on parallel architectures in order to tackle large scale geometries including a detailed description of the physics. Last decade, significant efforts have been devoted towards efficient implementation of the finite-differences methods on emerging architectures. These contributions have demonstrated their efficiency leading to robust industrial applications. The growing representation of heterogeneous architectures combining general purpose multicore platforms and accelerators leads to redesign current parallel application. In this paper, we consider StarPU task-based runtime system in order to harness the power of heterogeneous CPU+GPU computing nodes. We detail our implementation and compare the performance obtained with the classical CPU or GPU only versions. Preliminary results demonstrate significant speedups in comparison with the best implementation suitable for homogeneous cores.
Type de document :
Communication dans un congrès
27th International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD), Oct 2015, Florianopolis, Brazil. IEEE, 2015, IEEE 27th International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD). 〈http://www2.sbc.org.br/sbac/2015/〉. 〈10.1109/SBAC-PAD.2015.33〉
Liste complète des métadonnées

Littérature citée [26 références]  Voir  Masquer  Télécharger
https://hal.inria.fr/hal-01182746
Contributeur : Samuel Thibault <>
Soumis le : mercredi 5 août 2015 - 14:38:24
Dernière modification le : mardi 12 février 2019 - 21:04:52
Document(s) archivé(s) le : mercredi 26 avril 2017 - 08:21:21

Fichier

sbac2015_soumission.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

INRIA-BORDEAUX | UNIV-BORDEAUX | LABRI | INRIA | BRGM | PLAFRIM

Citation

Víctor Martínez, David Michéa, Fabrice Dupros, Olivier Aumage, Samuel Thibault, et al.. Towards seismic wave modeling on heterogeneous many-core architectures using task-based runtime system. 27th International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD), Oct 2015, Florianopolis, Brazil. IEEE, 2015, IEEE 27th International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD). 〈http://www2.sbc.org.br/sbac/2015/〉. 〈10.1109/SBAC-PAD.2015.33〉. 〈hal-01182746〉

Exporter

BibTeX TEI DC DCterms EndNote

Partager

Métriques

Consultations de la notice

605

Téléchargements de fichiers

458

Contact


archive-ouverte@inria.fr