Scheduling Moldable Tasks for Dynamic {SMP} Clusters in {S}o{C} Technology

Abstract : The paper presents an algorithm for scheduling parallel programs for execution in a parallel architecture based on dynamic SMP processor clusters with data transfers on the fly. The algorithm is based on the concept of moldable computational tasks. First, an initial program graph is decomposed into sub­graphs, which are then treated as moldable tasks. So identified moldable tasks are then scheduled using an algorithm with warranted schedule length.
Type de document :
Communication dans un congrès
Parallel Processing and Applied Mathematics, Sep 2005, Poznan, Poland, Springer-Verlag, 2005, Lecture Notes in Computer Science
Liste complète des métadonnées

Littérature citée [8 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/inria-00001079
Contributeur : Pierre-François Dutot <>
Soumis le : mercredi 1 février 2006 - 18:38:17
Dernière modification le : vendredi 21 décembre 2018 - 10:46:07
Document(s) archivé(s) le : samedi 3 avril 2010 - 22:02:24

Identifiants

  • HAL Id : inria-00001079, version 1

Collections

INRIA | IMAG | UGA

Citation

Lukasz Masko, Pierre-François Dutot, Grégory Mounié, Denis Trystram, Marek Tudruj. Scheduling Moldable Tasks for Dynamic {SMP} Clusters in {S}o{C} Technology. Parallel Processing and Applied Mathematics, Sep 2005, Poznan, Poland, Springer-Verlag, 2005, Lecture Notes in Computer Science. 〈inria-00001079〉

Partager

Métriques

Consultations de la notice

318

Téléchargements de fichiers

193