NewMadeleine: a Fast Communication Scheduling Engine for High Performance Networks

Olivier Aumage 1, 2 Elisabeth Brunet 1, 2 Nathalie Furmento 1, 2 Raymond Namyst 1, 2
1 RUNTIME - Efficient runtime systems for parallel architectures
INRIA Futurs, 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 : Communication libraries have dramatically made progress over the fifteen years, pushed by the success of cluster architectures as the preferred platform for high performance distributed computing. However, many potential optimizations are left unexplored in the process of mapping application communication requests onto low level network commands. The fundamental cause of this situation is that the design of communication subsystems is mostly focused on reducing the latency by shortening the critical path. In this paper, we present a new communication scheduling engine which dynamically optimizes application requests in accordance with the NICs capabilities and activity. The optimizing code is generic and portable. The database of optimizing strategies may be dynamically extended.
Keywords : NewMadeleine
Complete list of metadatas

Cited literature [11 references]  Display  Hide  Download

https://hal.inria.fr/inria-00127356
Contributor : Nathalie Furmento <>
Submitted on : Monday, January 29, 2007 - 2:57:32 PM
Last modification on : Tuesday, June 11, 2019 - 11:08:11 AM
Long-term archiving on : Friday, May 13, 2011 - 9:18:47 PM

File

cac-final.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00127356, version 1

Collections

Citation

Olivier Aumage, Elisabeth Brunet, Nathalie Furmento, Raymond Namyst. NewMadeleine: a Fast Communication Scheduling Engine for High Performance Networks. Workshop on Communication Architecture for Clusters (CAC 2007), workshop held in conjunction with IPDPS 2007, Mar 2007, Long Beach, California, United States. ⟨inria-00127356⟩

Share

Metrics

Record views

389

Files downloads

520