Parallelization, Distribution and Scaling of Multi-Simulations on Multi-Core Clusters, with DACCOSIM Environment - UMI 2958 - Axe de recherche : Computer Science Accéder directement au contenu
Autre Publication Scientifique Année : 2016

Parallelization, Distribution and Scaling of Multi-Simulations on Multi-Core Clusters, with DACCOSIM Environment

Résumé

DACCOSIM is a multi-simulation environment for continuous time systems , relying on FMI-CS standard and making easy the design of a multi-simulation graph. Moreover, DACCOSIM has been specially developed to run on multi-core PC clusters, in order to achieve speedup and size up on large scale multi-simulations. However, the distribution of the multi-simulation graph remains complex and is still under responsibility of the simulation developer. This paper introduces DACCOSIM parallel and distributed architecture, and our strategies to achieve efficient distribution of multi-simulation graph on multi-core clusters. Some performance experiments on two clusters, running up to 81 active components (FMU) and using up to 16 multi-core computing nodes, are shown. Performances measured on our faster cluster exhibit a good scalability, but some limitations of current DACCOSIM implementation are discussed.
Fichier principal
Vignette du fichier
TR-DACCOSIM-Scaling.pdf (1.48 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01289194 , version 1 (16-03-2016)
hal-01289194 , version 2 (17-03-2016)
hal-01289194 , version 3 (23-03-2016)

Licence

Copyright (Tous droits réservés)

Identifiants

  • HAL Id : hal-01289194 , version 3

Citer

Cherifa Dad, Stéphane Vialle, Mathieu Caujolle, Jean-Philippe Tavella, Michel Ianotto. Parallelization, Distribution and Scaling of Multi-Simulations on Multi-Core Clusters, with DACCOSIM Environment. 2016. ⟨hal-01289194v3⟩
229 Consultations
418 Téléchargements

Partager

Gmail Facebook X LinkedIn More