Hybrid scheduling for the parallel solution of linear systems

Jean-Yves L'Excellent 1 Patrick R. Amestoy 2 Stéphane Pralet 2 Abdou Guermouche 3, 4
1 GRAAL - Algorithms and Scheduling for Distributed Heterogeneous Platforms
Inria Grenoble - Rhône-Alpes, LIP - Laboratoire de l'Informatique du Parallélisme
4 SCALAPPLIX - Algorithms and high performance computing for grand challenge applications
INRIA Futurs, Université Bordeaux Segalen - Bordeaux 2, Université Sciences et Technologies - Bordeaux 1, École Nationale Supérieure d'Électronique, Informatique et Radiocommunications de Bordeaux (ENSEIRB), CNRS - Centre National de la Recherche Scientifique : UMR5800
Abstract : We consider the problem of designing a dynamic scheduling strategy that takes into account both workload and memory information in the context of the parallel multifrontal factorization. The originality of our approach is that we base our estimations (work and memory) on a static optimistic scenario during the analysis phase. This scenario is then used during the factorization phase to constrain the dynamic decisions that compute fully irregular partitions in order to better balance the workload. We show that our new scheduling algorithm significantly improves both the memory behaviour and the factorization time. We give experimental results for large challenging real-life 3D problems on 64 and 128 processors.
Type de document :
Article dans une revue
Parallel Computing, Elsevier, 2006, 32 (2), pp.136-156
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-00358623
Contributeur : Abdou Guermouche <>
Soumis le : mercredi 4 février 2009 - 03:15:36
Dernière modification le : mercredi 12 septembre 2018 - 17:46:02

Identifiants

  • HAL Id : hal-00358623, version 1

Citation

Jean-Yves L'Excellent, Patrick R. Amestoy, Stéphane Pralet, Abdou Guermouche. Hybrid scheduling for the parallel solution of linear systems. Parallel Computing, Elsevier, 2006, 32 (2), pp.136-156. 〈hal-00358623〉

Partager

Métriques

Consultations de la notice

525