A probabilistic model to analyse workflow performance on production grids

Tristan Glatard (1) , Johan Montagnat (1) , Xavier Pennec (2)

1 Laboratoire d'Informatique, Signaux, et Systèmes de Sophia-Antipolis (I3S) / Equipe MODALIS
2 ASCLEPIOS - Analysis and Simulation of Biomedical Images

Production grids are complex and highly variable systems whose behavior is not well understood and difficult to anticipate. The goal of this study is to estimate the impact of the variability of those infrastructures on the performance of workflow-based applications. A probabilistic model of workflows execution time is proposed and evaluated. Results show that the variability of the EGEE grid infrastructure impacts the execution time of a particular medical image analysis application by a factor 2. The model gives interesting insights on the grid behavior for different application parallelization modes.

Domaines: Calcul parallèle, distribué et partagé [cs.DC]

Tristan Glatard, Johan Montagnat, Xavier Pennec. A probabilistic model to analyse workflow performance on production grids. 8th IEEE International Symposium on Cluster Computing and the Grid (CCGRID'08), May 2008, Lyon, France. pp.510-517, ⟨10.1109/CCGRID.2008.123⟩. ⟨hal-00481713⟩