Cluster Computing and the Grid, IEEE International Symposium on (2008)
May 19, 2008 to May 22, 2008
Collective operations and non-blocking point-to-point operations are two important parts of MPI that each provide important performance and programmability benefits. Although non-blocking collective operations are an obvious extension to MPI, there have been no comprehensive studies of this functionality. This dissertation will study non-blocking collective operations, integrating theory, practice, and application. We use a well-understood network model to found our theoretical analyses and we realize our communication operations as a portable library layered on MPI. A real-world quantum-mechanical application is used as a deployment and evaluation vehicle for our approach.
Overlap, MPI, non-blocking collective communication, program optimization
A. Lumsdaine and T. Hoefler, "Overlapping Communication and Computation with High Level Communication Routines," 2008 8th International Symposium on Cluster Computing and the Grid (CCGRID '08)(CCGRID), Lyon, 2008, pp. 572-577.