Analyse et déploiement de solutions algorithmiques et logicielles pour des applications bioinformatiques à grande échelle sur la grille

Raphaël Bolze 1, 2
1 GRAAL - Algorithms and Scheduling for Distributed Heterogeneous Platforms
Inria Grenoble - Rhône-Alpes, LIP - Laboratoire de l'Informatique du Parallélisme
Abstract : This thesis was conducted by the needs of the Decrypthon project (collaborative project between AFM, CNRS and IBM). First we show the role of architect played in order to select and define the Decrypthon grid infrastructure. The resources of this grid are hosted by five Universities (Bordeaux I, Lille I, ENS-Lyon, Pierre et Marie Curie Paris VI et Orsay). The network connexion is provided by RENATER (Réseau National de Télécommunications pour l'Enseignement et la Recherche). The CRIHAN ( Centre de ressources Informatiques de Hautes Normandie) is also involved into this parternship and provides data warehouse for scientists. In a second hand we present several experiments carried on Grid'5000 in order to validate the grid middleware DIET and its tools on a large scale platform such as Grid'5000. On this research platform, we also studied the application of the project "Help Cure Muscular Dystrophy", one of the project selected by the Decrypthon. This study prepared the launch of a 6 months computing phase on the volunteers grid : World Community Grid support by IBM US. The document presents all steps before and after the computing phase which require more than 80 centuries of CPU time on the volunteers device. Finally, we have designed several heuristics to tackle the problem of online multi-workflow scheduling in a shared grid environment. We have implemented those heuristics into DIET middleware and we have validated their behavior with case study applications from Decrypthon. This work required many software developments in the aim to grid enabled bioinformatic applications and transparenlty give access to the Decrypthon grid, but also into DIET middleware and tools around : DIET_Webboard, VizDIET, GoDIET, LogService, MA_DAG, etc. The results exposed in this thesis were obtained with tree different grids : the Decrypthon grid, the volunteer grid (World Community Grid) and the research grid (Grid'5000).
Complete list of metadatas

Cited literature [93 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-00344249
Contributor : Raphaël Bolze <>
Submitted on : Thursday, December 4, 2008 - 11:45:23 AM
Last modification on : Saturday, April 21, 2018 - 1:27:20 AM
Long-term archiving on : Thursday, October 11, 2012 - 12:31:44 PM

Identifiers

  • HAL Id : tel-00344249, version 1

Citation

Raphaël Bolze. Analyse et déploiement de solutions algorithmiques et logicielles pour des applications bioinformatiques à grande échelle sur la grille. Modélisation et simulation. Ecole normale supérieure de lyon - ENS LYON, 2008. Français. ⟨tel-00344249⟩

Share

Metrics

Record views

642

Files downloads

1179