Evaluation et optimisation des performances d'une implementation de MPI pour la grille

The MPI standard is often used in parallel applications for communication needs. Most of them are designed for homogeneous clusters but MPI implementations for grids have to take into account heterogeneity and long distance network links in order to maintain a high performance level. These two constraints are not considered together in existing MPI implementations. State of the art of MPI implementations is followed by analysis of one of them, MPICH-Madeleine, which manages heterogeneity with efficiency. Our optimizations lead to obtain a bandwith of 300 Mbps instead of 40 Mbps for sending MPI messages over a Wide Area Network. The experiments have been performed on the Grid’5000 french national plateform.