On using connection-oriented vs. connection-less transport for performance and scalability of collective and one-sided operations
Authors:
Open Access
- 14 March 2007
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 46-54
Abstract
No abstract availableThis publication has 9 references indexed in Scilit:
- Breaking the Connection: RDMA Deconstructed2006
- Efficient Shared Memory and RDMA Based Design for MPI_Allgather over InfiniBand2006
- Supporting MPI-2 One Sided Communication on Multi-rail InfiniBand Clusters: Design Challenges and Performance BenefitsLecture Notes in Computer Science, 2005
- Communication optimizations for fine-grained UPC applications2005
- HUNTing the overlap2005
- Efficient Implementation of MPI-2 Passive One-Sided Communication on InfiniBand ClustersLecture Notes in Computer Science, 2004
- A performance analysis of the Berkeley UPC compiler2003
- Efficient algorithms for all-to-all communications in multiport message-passing systemsIEEE Transactions on Parallel and Distributed Systems, 1997
- Users guide for mpich, a portable implementation of MPI1996