Data Sharing Mechanisms for Parallel Graph Algorithms on the Intel SCC

Abstract : On many-core processors that do not provide hard-ware cache coherence, using shared memory in parallel computations is challenging. Reverting to pure message passing would avoid consistency issues, but replicating large shared datasets by messages is less efficient than accessing them directly through shared memory. The TACO-MESH framework provides lightweight remote method calls and shared objects with software-managed consistency. This paper presents experience from porting a graph partitioning algorithm to the framework. A performance evaluation on the experimental Intel SCC processor, which has no hardware cache coherence, shows that parallelization can be efficient despite the overhead of software-level consistency management.
Keywords : Many-Core Intel SCC
Type de document :
Communication dans un congrès
Eric Noulard and Simon Vernhes. The 6th Many-core Applications Research Community (MARC) Symposium, Jul 2012, Toulouse, France. ONERA, The French Aerospace Lab, pp.13-18, 2012
Liste complète des métadonnées

Littérature citée [23 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-00718993
Contributeur : Simon Vernhes <>
Soumis le : mercredi 18 juillet 2012 - 16:31:44
Dernière modification le : mercredi 18 juillet 2012 - 17:21:13
Document(s) archivé(s) le : vendredi 19 octobre 2012 - 03:00:49

Fichier

MARC6_Data-Sharing-Mechanisms-...
Accord explicite pour ce dépôt

Identifiants

  • HAL Id : hal-00718993, version 1

Collections

Citation

Randolf Rotta, Thomas Prescher, Jana Traue, Jörg Nolte. Data Sharing Mechanisms for Parallel Graph Algorithms on the Intel SCC. Eric Noulard and Simon Vernhes. The 6th Many-core Applications Research Community (MARC) Symposium, Jul 2012, Toulouse, France. ONERA, The French Aerospace Lab, pp.13-18, 2012. 〈hal-00718993〉

Partager

Métriques

Consultations de la notice

341

Téléchargements de fichiers

315