A sampling-based approach for communication libraries auto-tuning
| hal.structure.identifier | Département Informatique [TSP - INF] | |
| dc.contributor.author | BRUNET, Elisabeth | |
| hal.structure.identifier | Département Informatique [TSP - INF] | |
| dc.contributor.author | TRAHAY, François | |
| hal.structure.identifier | Efficient runtime systems for parallel architectures [RUNTIME] | |
| hal.structure.identifier | Laboratoire Bordelais de Recherche en Informatique [LaBRI] | |
| dc.contributor.author | DENIS, Alexandre | |
| hal.structure.identifier | Efficient runtime systems for parallel architectures [RUNTIME] | |
| hal.structure.identifier | Laboratoire Bordelais de Recherche en Informatique [LaBRI] | |
| dc.contributor.author | NAMYST, Raymond | |
| dc.date.accessioned | 2024-04-15T09:47:15Z | |
| dc.date.available | 2024-04-15T09:47:15Z | |
| dc.date.issued | 2011-09-26 | |
| dc.date.conference | 2011-09-26 | |
| dc.identifier.uri | https://oskar-bordeaux.fr/handle/20.500.12278/198082 | |
| dc.description.abstractEn | 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. | |
| dc.description.sponsorship | Gestion de ressources coopérative multi niveaux - ANR-09-COSI-0001 | |
| dc.language.iso | en | |
| dc.subject.en | NewMadeleine | |
| dc.subject.en | MadMPI | |
| dc.subject.en | MPI | |
| dc.title.en | A sampling-based approach for communication libraries auto-tuning | |
| dc.type | Communication dans un congrès | |
| dc.subject.hal | Informatique [cs]/Réseaux et télécommunications [cs.NI] | |
| bordeaux.hal.laboratories | Laboratoire Bordelais de Recherche en Informatique (LaBRI) - UMR 5800 | * |
| bordeaux.institution | Université de Bordeaux | |
| bordeaux.institution | Bordeaux INP | |
| bordeaux.institution | CNRS | |
| bordeaux.conference.title | IEEE International Conference on Cluster Computing | |
| bordeaux.country | US | |
| bordeaux.conference.city | Austin | |
| bordeaux.peerReviewed | oui | |
| hal.identifier | inria-00605735 | |
| hal.version | 1 | |
| hal.invited | non | |
| hal.proceedings | oui | |
| hal.conference.end | 2011-09-30 | |
| hal.popular | non | |
| hal.audience | Internationale | |
| hal.origin.link | https://hal.archives-ouvertes.fr//inria-00605735v1 | |
| bordeaux.COinS | ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.date=2011-09-26&rft.au=BRUNET,%20Elisabeth&TRAHAY,%20Fran%C3%A7ois&DENIS,%20Alexandre&NAMYST,%20Raymond&rft.genre=unknown |
Fichier(s) constituant ce document
| Fichiers | Taille | Format | Vue |
|---|---|---|---|
|
Il n'y a pas de fichiers associés à ce document. |
|||