Identifying the capability of overlapping computation with communication

A. Sohn*, J. Ku, Y. Kodama, M. Sato, H. Sakane, H. Yamana, S. Sakai, Y. Yamaguchi

*Corresponding author for this work

Research output: Contribution to journalConference articlepeer-review


Overlapping computation with communication is central to obtaining high performance on distributed-memory multiprocessors. This report explicates the overlapping capability of two distributed-memory multiprocessors: the EM-X and IBM SP-2. The well-known bitonic sorting algorithm is selected for experiments. Various message sizes are used to determine when, where, how much and why overlapping takes place. Experimental results indicate that both multiprocessors would yield up to 30% to 40% overlap of communication time when the message size is approximately 1K integers. EM-X is found message-size insensitive yielding high overlap for various message sizes while SP-2 was effective for the window of message size 512 to 2K integers.

Original languageEnglish
Pages (from-to)133-138
Number of pages6
JournalParallel Architectures and Compilation Techniques - Conference Proceedings, PACT
Publication statusPublished - 1996 Jan 1
Externally publishedYes
EventProceedings of the 1996 Conference on Parallel Architectures and Compilation Techniques, PACT'96 - Boston, MA, USA
Duration: 1996 Oct 201996 Oct 23

ASJC Scopus subject areas

  • Software
  • Theoretical Computer Science
  • Hardware and Architecture


Dive into the research topics of 'Identifying the capability of overlapping computation with communication'. Together they form a unique fingerprint.

Cite this