TY - GEN
T1 - MPI applications on grids
T2 - Euro-Par 2009 Parallel Processing - 15th International Euro-Par Conference, Proceedings
AU - Coti, Camille
AU - Herault, Thomas
AU - Cappello, Franck
PY - 2009
Y1 - 2009
N2 - Porting on grids complex MPI applications involving collective communications requires significant program modification, usually dedicated to a single grid structure. The difficulty comes from the mismatch between programs organizations and grid structures: 1) large grids are hierarchical structures aggregating parallel machines through an interconnection network, decided at runtime and 2) the MPI standard does not currently provide any specific information for topology-aware applications, so almost all MPI applications have been developed following a non-hierarchical and non-flexible vision. In this paper, we propose a generic programming method and a modification of the MPI runtime environment to make MPI applications topology aware. In contrary to previous approaches, topology requirements for the application are given to the grid scheduling system, which exposes the compatible allocated topology to the application.
AB - Porting on grids complex MPI applications involving collective communications requires significant program modification, usually dedicated to a single grid structure. The difficulty comes from the mismatch between programs organizations and grid structures: 1) large grids are hierarchical structures aggregating parallel machines through an interconnection network, decided at runtime and 2) the MPI standard does not currently provide any specific information for topology-aware applications, so almost all MPI applications have been developed following a non-hierarchical and non-flexible vision. In this paper, we propose a generic programming method and a modification of the MPI runtime environment to make MPI applications topology aware. In contrary to previous approaches, topology requirements for the application are given to the grid scheduling system, which exposes the compatible allocated topology to the application.
UR - http://www.scopus.com/inward/record.url?scp=70350641510&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=70350641510&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-03869-3_45
DO - 10.1007/978-3-642-03869-3_45
M3 - Conference contribution
AN - SCOPUS:70350641510
SN - 3642038689
SN - 9783642038686
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 466
EP - 477
BT - Euro-Par 2009 Parallel Processing - 15th International Euro-Par Conference, Proceedings
Y2 - 25 August 2009 through 28 August 2009
ER -