Today, clusters are often interconnected by long distance networks within
grids to offer a huge number of available ressources to a range of users. MPI,
the standard communication library used to write parallel applications,
has been implemented for clusters. Two main features of grids: long distance
networks and technological heterogeneity, raise the question of MPI efficiency
in grids.
This work evaluate four recent MPI implementations (MPICH2, MPICH-Madeleine,
OpenMPI and GridMPI) in the french
research grid: Grid'5000. The comparison is based on the execution of pingpong,
NAS Parallel Benchmarks and a real application in geophysics (ray2mesh). We show that this
implementations present performance differences. Executing MPI applications on
the grid can be beneficial if the parameters are well tuned.
Collective operations have to be improved to be efficient.
Publications:
Comparison and tuning of MPI implementation in a grid context
Ludovic Hablot, Olivier Glück, Jean-Christophe Mignot, Stéphane Genaud et Pascale Vicat-Blanc Primet
Poster paper, in Proceedings of 2007 IEEE International Conference on Cluster Computing (CLUSTER), pages 458-463, Septembre 2007.
[PDF][Poster][Bibtex]
Research report : Comparison and tuning of MPI implementation in a grid context
Ludovic Hablot, Olivier Glück, Jean-Christophe Mignot, Stéphane Genaud et Pascale Vicat-Blanc Primet
RR INRIA 6200, Juillet 2007.
[PDF][Bibtex]