Skip to Main content Skip to Navigation

Athapascan-1 : interface générique pour l'ordonnancement dans un environnement d'exécution parallèle

Abstract : In a parallel programming environment, the load sharing module - or application level scheduler - manages the execution of a program over the resources of a distributed architecture: processors and memory modules. In practice, the efficiency of the scheduling strategy highly depends on the characteristics of both application and target architecture. Consequently, to achieve code portability, it should be possible to tune the scheduling strategy. In this thesis, we propose the use of a dynamic data flow graph as the key element in order to separate the application code from its scheduling strategy. This approach allows the construction of a modular and generic software environment that enables to tune the scheduling strategy. The specification of this environment relies on the identification of interfaces between the load sharing module and both the application and the architecture. This approach, based on a macroscopic analysis of the data flow graph, is original: we show that it can be seen as an extension of other traditional load balancing systems. Finally, an experimental validation of this environment was carried out through its implementation in Athapascan-1. Athapascan-1 is the programming interface of the Athapascan parallel programming environment developed in the APACHE project. Various scheduling strategies (static, dynamic and hybrid) have been implemented. We present the performance of some of these strategies applied to several programs on various architectures.
Complete list of metadatas

Cited literature [47 references]  Display  Hide  Download
Contributor : Thèses Imag <>
Submitted on : Wednesday, February 18, 2004 - 11:40:02 AM
Last modification on : Friday, November 6, 2020 - 4:39:25 AM
Long-term archiving on: : Friday, April 2, 2010 - 8:26:49 PM


  • HAL Id : tel-00004816, version 1




Gerson Geraldo Homrich Cavalheiro. Athapascan-1 : interface générique pour l'ordonnancement dans un environnement d'exécution parallèle. Réseaux et télécommunications [cs.NI]. Institut National Polytechnique de Grenoble - INPG, 1999. Français. ⟨tel-00004816⟩



Record views


Files downloads