Static/Dynamic Validation of MPI Collective Communications in Multi-threaded Context

Abstract : Scientific applications mainly rely on the MPI parallel programming model to reach high performance on supercomputers. The advent of manycore architectures (larger number of cores and lower amount of memory per core) leads to mix MPI with a thread-based model like OpenMP. But integrating two different programming models inside the same application can be tricky and generate complex bugs. Thus, the correctness of hybrid programs requires a special care regarding MPI calls location. For example, identical MPI collective operations cannot be performed by multiple non-synchronized threads. To tackle this issue, this paper proposes a static analysis and a reduced dynamic instrumentation to detect bugs related to misuse of MPI collective operations inside or outside threaded regions. This work extends PARCOACH [4] designed for MPI-only applications and keeps the compatibility with these algorithms. We validated our method on multiple hybrid benchmarks and applications with a low overhead.
Type de document :
Communication dans un congrès
the 20th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, Feb 2015, San Francisco, United States. Proceedings of the 20th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, 50, pp.279-280, 〈10.1145/2688500.2688548〉
Liste complète des métadonnées

Littérature citée [4 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-01253204
Contributeur : Emmanuelle Saillard <>
Soumis le : lundi 11 janvier 2016 - 16:59:46
Dernière modification le : jeudi 29 mars 2018 - 09:04:04
Document(s) archivé(s) le : mardi 12 avril 2016 - 11:01:18

Fichier

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

Identifiants

Collections

Citation

Emmanuelle Saillard, Patrick Carribault Cea, Denis Barthou. Static/Dynamic Validation of MPI Collective Communications in Multi-threaded Context. the 20th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, Feb 2015, San Francisco, United States. Proceedings of the 20th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, 50, pp.279-280, 〈10.1145/2688500.2688548〉. 〈hal-01253204〉

Partager

Métriques

Consultations de la notice

75

Téléchargements de fichiers

100