Recovering numerical reproducibility in hydrodynamic simulations

Philippe Langlois 1 Rafife Nheili 1 Christophe Denis 2
1 DALI - Digits, Architectures et Logiciels Informatiques
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier, UPVD - Université de Perpignan Via Domitia
Abstract : HPC simulations suffer from failures of numerical reproducibility because of floating-point arithmetic peculiarities. Different computing distributions of a parallel computation may yield different numerical results. We are interested in a finite element computation of hydrodynamic simulations within the openTelemac software where parallelism is provided by domain decomposition. One main task in a finite element simulation consists in building one large linear system and to solve it. Here the building step relies on element-by-element storage mode and the solving step applies the conjugated gradient algorithm. The subdomain parallelism is merged within these steps. We study why reproducibility fails in this process and which operations have to be corrected. We detail how to use compensation techniques to compute a numerically reproducible resolution. We illustrate this approach presenting the reproducible version of hydrodynamic simulations for one test cases provided with the openTelemac software suite.
Type de document :
Communication dans un congrès
ARITH: Computer Arithmetic, Jul 2016, Silicon Valley, Santa Clara, CA, United States. 23th IEEE International Symposium on Computer Arithmetic, 2016, <http://arith23.gforge.inria.fr/>
Liste complète des métadonnées


https://hal-lirmm.ccsd.cnrs.fr/lirmm-01274671
Contributeur : Rafife Nheili <>
Soumis le : mardi 16 février 2016 - 10:07:22
Dernière modification le : vendredi 9 juin 2017 - 10:43:02
Document(s) archivé(s) le : mardi 17 mai 2016 - 17:23:25

Fichier

LaNhDe_2016 (1).pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : lirmm-01274671, version 1

Citation

Philippe Langlois, Rafife Nheili, Christophe Denis. Recovering numerical reproducibility in hydrodynamic simulations. ARITH: Computer Arithmetic, Jul 2016, Silicon Valley, Santa Clara, CA, United States. 23th IEEE International Symposium on Computer Arithmetic, 2016, <http://arith23.gforge.inria.fr/>. <lirmm-01274671>

Partager

Métriques

Consultations de
la notice

176

Téléchargements du document

185