Implementation and Evaluation of Shared-Memory Communication and Synchronization Operations in MPICH2 using the Nemesis Communication Subsystem
MERCIER, Guillaume
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
Efficient runtime systems for parallel architectures [RUNTIME]
Ecole Nationale Supérieure d'Electronique, Informatique et Radiocommunications de Bordeaux [ENSEIRB]
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
Efficient runtime systems for parallel architectures [RUNTIME]
Ecole Nationale Supérieure d'Electronique, Informatique et Radiocommunications de Bordeaux [ENSEIRB]
MERCIER, Guillaume
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
Efficient runtime systems for parallel architectures [RUNTIME]
Ecole Nationale Supérieure d'Electronique, Informatique et Radiocommunications de Bordeaux [ENSEIRB]
< Réduire
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
Efficient runtime systems for parallel architectures [RUNTIME]
Ecole Nationale Supérieure d'Electronique, Informatique et Radiocommunications de Bordeaux [ENSEIRB]
Langue
en
Article de revue
Ce document a été publié dans
Parallel Computing. 2007-09, vol. 33, n° 9, p. 634-644
Elsevier
Résumé en anglais
This paper presents the implementation of MPICH2 over the Nemesis communication subsystem and the evaluation of its shared-memory performance. We describe design issues as well as some of the optimization techniques we ...Lire la suite >
This paper presents the implementation of MPICH2 over the Nemesis communication subsystem and the evaluation of its shared-memory performance. We describe design issues as well as some of the optimization techniques we employed. We conducted a performance evaluation over shared memory using microbenchmarks. The evaluation shows that MPICH2 Nemesis has very low communication overhead, making it suitable for smaller-grained applications.< Réduire
Mots clés en anglais
MPI
shared memory
parallel programming
Origine
Importé de halUnités de recherche