2006 International Conference on Parallel Processing (ICPP'06) Data Transfers between Processes in an SMP System: Performance Study and Application to MPI Columbus, Ohio August 14-August 18 ISBN: 0-7695-2636-5
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/ICPP.2006.31
This paper focuses on the transfer of large data in SMP systems. Achieving good performance for intranode communication is critical for developing an efficient communication system, especially in the context of SMP clusters. We evaluate the performance of five transfer mechanisms: sharedmemory buffers, message queues, the Ptrace system call, kernel module-based copy, and a high-speed network. We evaluate each mechanism based on latency, bandwidth, its impact on application cache usage, and its suitability to support MPI twosided and one-sided messages.
Citation:
Darius Buntinas, Guillaume Mercier, William Gropp, "Data Transfers between Processes in an SMP System: Performance Study and Application to MPI," icpp, pp.487-496, 2006 International Conference on Parallel Processing (ICPP'06), 2006 Usage of this product signifies your acceptance of the Terms of Use. | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||