Communication operations on coarse-grained mesh architectures
Parallel Computing
MagPIe: MPI's collective communication operations for clustered wide area systems
Proceedings of the seventh ACM SIGPLAN symposium on Principles and practice of parallel programming
How helpers hasten h-relations
Journal of Algorithms
MPI-The Complete Reference, Volume 1: The MPI Core
MPI-The Complete Reference, Volume 1: The MPI Core
The Hierarchical Factor Algorithm for All-to-All Communication (Research Note)
Euro-Par '02 Proceedings of the 8th International Euro-Par Conference on Parallel Processing
MPI Optimization for SMP Based Clusters Interconnected with SCI
Proceedings of the 7th European PVM/MPI Users' Group Meeting on Recent Advances in Parallel Virtual Machine and Message Passing Interface
MPI-2 One-Sided Communications on a Giganet SMP Cluster
Proceedings of the 8th European PVM/MPI Users' Group Meeting on Recent Advances in Parallel Virtual Machine and Message Passing Interface
Bandwidth-Efficient Collective Communication for Clustered Wide Area Systems
IPDPS '00 Proceedings of the 14th International Symposium on Parallel and Distributed Processing
Exploiting Hierarchy in Parallel Computer Networks to Optimize Collective Operation Performance
IPDPS '00 Proceedings of the 14th International Symposium on Parallel and Distributed Processing
Collective operations in NEC's high-performance MPI libraries
IPDPS'06 Proceedings of the 20th international conference on Parallel and distributed processing
| Hi-index | 0.00 |
We present the implementation of an improved, almost optimal algorithm for regular, personalized all-to-all communication for hierarchical multiprocessors, like clusters of SMP nodes. In MPI this communication primitive is realized in the MPI_Alltoall collective. The algorithm is a natural generalization of a well-known algorithm for nonhierarchical systems based on factorization. A specific contribution of the paper is a completely contention-free scheme not using token-passing for exchange of messages between SMP nodes.We describe a dedicated implementation for a small Giganet SMP cluster with 6 SMP nodes of 4 processors each. We present simple experiments to validate the assumptions underlying the design of the algorithm. The results were used to guide the detailed implementation of a crucial part of the algorithm. Finally, we compare the improved MPI_Alltoall collective to a trivial (but widely used) implementation, and show improvements in average completion time of sometimes more than 10%. While this may not seem much, we have reasons to believe that the improvements will be more substantial for larger systems.