BIP-SMP : High Performance Message Passing over a Cluster of Commodity SMPs

  title={BIP-SMP : High Performance Message Passing over a Cluster of Commodity SMPs},
  author={Patrick Geoffray and Lo{\"i}c Prylli and Bernard Tourancheau},
  journal={ACM/IEEE SC 1999 Conference (SC'99)},
Device Interface Channel Interface NX Check_incoming "short", "eager", P4 TCP/IP Paragon SP/2 Generic ADI code, datatype mgmt, heterogeneity request queues mgmt " Protocol interface" SGI port. other ports shared-mem port MPL BIP MPI BIP "rendez-vous" Protocols Figure1: Thearchitectureof MPI-BIP implementedwith oneor several messagesof the underlyingcommunicationsystem (BIP in our case). Thecostof MPI-BIP is approximatelyanoverheadof 2 s (mainlyCPU)overBIP for thelatency onourcluster. Thus… CONTINUE READING
31 Citations
3 References
Similar Papers


Publications citing this paper.
Showing 1-10 of 31 extracted citations


Publications referenced by this paper.
Showing 1-3 of 3 references

University of Tennessee / dongar

  • Jack Dongarra
  • Linpack benchmark-parallel, April
  • 1999
Highly Influential
3 Excerpts

ActiveMessages:an Efficient CommunicationArchitecture for Multiprocessors

  • T. von Eicken
  • PhD thesis,University of California at Berkeley,
  • 1993
Highly Influential
4 Excerpts

Similar Papers

Loading similar papers…