Grid'5000 user report for

Jump to: navigation, search

User information

( user)
More user information in the user management interface.

Experiments

  • LogCentral :Scalabity test and Validation (Grid monitoring) (Middleware) [achieved]
    Description: We have tested across the grid5000 clusters the behavior of our monitoring system for DIET : LogCentral. We have also valdiated the model of communication between LogCentral and DIET component and the monitoring tool
    Results:
    More information here
  • LogService impact in DIET (Grid monitoring) (Middleware) [achieved]
    Description: In this experiment we have taken care of the impact of the LogService ( monitoring system) for DIET in the case of a large number of requests.
    Results:
    More information here
  • Scalability of DIET (Middleware) [achieved]
    Description: We have tested the scalabilty of DIET. We have launched DIET over 9 sites and use more than 600 nodes in order to provide a prove of testing and add some feedback about performance of DIET.
    Results:
    illustrating chart picture not found
    More information here
  • DIET experiment (Middleware) [in progress]
    Description: Many experiment to do a large scale experiment for the DIET middleware (I give more details with to examples in the two following experiments). The aim of the DIET project is to develop a set of tools to build computational servers. Huge problems can now be computed over the Internet thanks to Grid Computing Environments like Globus or Legion. Because most of current applications are numerical, the use of libraries like BLAS, LAPACK, ScaLAPACK or PETSc is mandatory. The integration of such libraries in high level applications using languages like Fortran or C is far from being easy. Moreover, the computational power and memory needs of such applications may of course not be available on every workstation. Thus, the RPC seems to be a good candidate to build Problem Solving Environments on the Grid. Several tools following this approach exist, like Netsolve, NINF, NEOS, or RCS.
    Results: Some results about DIET scheduling performance when using 7 differents sites
    illustrating chart picture not found
    More information here
  • Decrypthon project (BioInformatics) (Application) [achieved]
    Description: I have used g5k to launch a set of 28224 x 2 jobs to evaluate time taken for a docking program During this test i use about 4 sites and about 640 nodes. This test consume about 73 j 2 h 41 s and 98 j 10 h 9 min 44 s cpu time
    Results: I know now that the program will take 14 centruies 88 years 237 days 19 hours 45 minutes 54 seconds and 19 centruies 79 years 305 days 9 hours 26 minutes 48 seconds
    illustrating chart picture not found
    More information here
  • DIET experiment with workflow scheduling heuristics (Middleware) [in progress]
    Description: 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 and ANR GWENDIA.
    Results:
  • DIET experiment with workflow scheduling heuristics (Middleware) [in progress]
    Description: 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 and ANR GWENDIA.
    Results:
  • Experiments with DIET/autodock (Application) [in progress]
    Description: In silico drug discovery is one of the most promising strategies to speed-up the drug development process. The propose of this experiment was to prototype the application (autodock) within the DIET grid middleware and test it in a large scale infrastructure such as Grid'5000. This work is done in the framework of the ANR GWENDIA.
    Results: This experiment was a success and validate the prototype of the application. The application is grid enabled with DIET.
  • Experiments with DIET/cardiac imaging (Application) [in progress]
    Description: Cardio-vascular diseases are the leading cause of mortality in our countries. Medical imaging techniques have considerably evolved and now offer a privileged way for studying pathological vessels and hearts. The aim of this experiment is to use workflows to peed up the design of new cardio-vascular image processing chains and to ease setting up evaluation protocols of such processes on large databases by combining processing tasks. The propose of this experiment was to prototype the application (cardiac imaging) within the DIET grid middleware by using workflow enactement feature of DIET. This work is done in the framework of the ANR GWENDIA.
    Results: This experiment was a success and validate the prototype of the application. The application is grid enabled with DIET and use workflow enactement feature of DIET.

Publications

  • Analyse et déploiement de solutions algorithmiques et logicielles pour des applications bioinformatiques à grande échelle sur la grille [2008] (national)
    EntryType: phdthesis
    Author: Bolze, Raphaël
    Month: Octobre
    Address: LIP, Lyon, France
    School: Université de Lyon - Ecole Normale Supérieure de Lyon
    Type: Thèse de doctorat
    Url: http://www.ens-lyon.fr/LIP/Pub/Rapports/PhD/PhD2008/PhD2008-09.pdf
    Keywords: DIET, intergiciel de grille, bioinformatique, ordonnancement multi-workflow, calcul sur grille de volontaire, étude compartive de grilles, Décrypthon
  • Large Scale Execution of a Bioinformatic Application on a Volunteer Grid [2008] (international)
    EntryType: inproceedings
    Author: Bertis, Viktor and Bolze, Raphaël and Desprez, Frédéric and Reed, Kevin
    Booktitle: Workshop on Parallel and Distributed Scientific and Engineering Computing (PDSEC), in conjuction with IPDPS
    Month: April
  • Grid'5000: a large scale and highly reconfigurable experimental Grid testbed. [2006] (international)
    EntryType: article
    Author: R. Bolze, F. Cappello, E. Caron, M. Daydé, F. Desprez, E. Jeannot, Y. Jégou, S. Lantéri, J. Leduc, N. Melab, G. Mornet,R. Namyst, P. Primet, B. Quetier, O. Richard, E-G. Talbi and I. Touche
    Journal: International Journal of High Performance Computing Applications
    Volume: 20
    Pages: 481-494
    Number: 4
    Month:
    Abstract: Large scale distributed systems like Grids are difficult to study only from theoretical models and simulators. Most Grids deployed at large scale are production platforms that are inappropriate research tools because of their limited reconfiguration, control and monitoring capabilities. In this paper, we present Grid?5000, a 5000 CPUs nation-wide infrastructure for research in Grid computing. Grid?5000 is designed to provide a scientific tool for computer scientists similar to the large-scale instruments used by physicists, astronomers, and biologists. We describe themotivations, design considerations, architecture, control, and monitoring infrastructure of this experimental platform. We present configuration examples and performance results for the reconfiguration subsystem.
    Keywords: Grid'5000

Collaborations

    Success stories and benefits from Grid'5000

    • Overall benefits
    • Wild access to a large number of computers.

    last update: 2008-12-11 12:59:34

    Personal tools
    Namespaces

    Variants
    Views
    Actions
    Public Portal
    Users Portal
    Admin portal
    Wiki special pages
    Toolbox