MPI Support for Multi-core Architectures: Optimized Shared Memory Collectives

Conference · Tue Jan 01 00:00:00 EST 2008

OSTI ID:1004675

Graham, Richard L ^[1]; Shipman, Galen M ^[1]

---

1. ORNL

With local core counts on the rise, taking advantage of shared memory to optimize collective operations can improve performance. We study several on-host shared memory optimized algorithms for MPI Bcast, MPI Reduce, and MPI Allreduce, using tree-based, and reduce-scatter algorithms. For small data operations with relatively large synchronization costs fan-in/fan-out algorithms generally perform best. For large messages data manipulation constitute the largest cost and reduce-scatter algorithms are best for reductions. These optimization improve performance by up to a factor of three. Memory and cache sharing effect require deliberate process layout and careful radix selection for tree-based methods

• Research Organization: Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States). National Center for Computational Sciences (NCCS)
• Sponsoring Organization: USDOE Office of Science (SC)
• DOE Contract Number: DE-AC05-00OR22725
• OSTI ID: 1004675
• Resource Relation: Conference: Proceedings of the 15th European PVM/MPI Users' Group Meeting on Recent Advances in Parallel Virtual Machine and Message Passing Interface, Dublin, Ireland, 20080907, 20080907
• Country of Publication: United States
• Language: English

Related Subjects

99 GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE
ALGORITHMS
OPTIMIZATION
PERFORMANCE
SYNCHRONIZATION