BRUNET, Elisabeth; TRAHAY, François; DENIS, Alexandre; NAMYST, Raymond

Métadonnées

Licence d’utilisation du document

BRUNET, Elisabeth
Département Informatique [TSP - INF]

TRAHAY, François
Département Informatique [TSP - INF]

DENIS, Alexandre
Efficient runtime systems for parallel architectures [RUNTIME]
Laboratoire Bordelais de Recherche en Informatique [LaBRI]

Langue

Communication dans un congrès

Ce document a été publié dans

IEEE International Conference on Cluster Computing, 2011-09-26, Austin. 2011-09-26

Résumé en anglais

Communication performance is a critical issue in HPC applications, and many solutions have been proposed on the literature (algorithmic, protocols, etc.) In the meantime, computing nodes become massively multicore, leading to a real imbalance between the number of communication sources and the number of physical communication resources. Thus it is now mandatory to share network boards between computation flows, and to take this sharing into account while performing communication optimizations. In previous papers, we have proposed a model and a framework for on-the-fly optimizations of multiplexed concurrent communication flows, and implemented this model in the \nm communication library. This library features optimization strategies able for example to aggregate several messages to reduce the number of packets emitted on the network, or to split messages to use several NICs at the same time. In this paper, we study the tuning of these dynamic optimization strategies. We show that some parameters and thresholds (\rdv threshold, aggregation packet size) depend on the actual hardware, both host and NICs. We propose and implement a method based on sampling of the actual hardware to auto-tune our strategies. Moreover, we show that multi-rail can greatly benefit from performance predictions. We propose an approach for multi-rail that dynamically balance the data between NICs using predictions based on sampling.< Réduire

Mots clés en anglais

NewMadeleine

MadMPI

MPI

URI

https://oskar-bordeaux.fr/handle/20.500.12278/198082

Project ANR

Gestion de ressources coopérative multi niveaux - ANR-09-COSI-0001

Origine

Importé de hal

Unités de recherche

Laboratoire Bordelais de Recherche en Informatique (LaBRI) - UMR 5800