Open MPI logo

Open MPI

  |   Home   |   Support   |   FAQ   |  

Title: A Case for Standard Non-Blocking Collective Operations

Author(s):

Torsten Hoefler, Prabhanjan Kambadur, Richard L. Graham, Galen Shipman, Andrew Lumsdaine

Abstract:

In this paper we make the case for adding standard non-blocking collective operations to the MPI standard. The non-blocking point-to-point and blocking collective operations currently defined by MPI provide important performance and abstraction benefits. To allow these benefits to be simultaneously realized, we present an application programming interface for non-blocking collective operations in MPI. Microbenchmark and application-based performance results demonstrate that non-blocking collective operations offer not only improved convenience, but improved performance as well, when compared to manual use of threads with blocking collectives.

Presented: EuroPVM/MPI '07, September 30th - October 3rd, 2007, Paris, France.

Paper:

mpi-vs-nbc.pdf (PDF)

Bibtex reference:

 @InProceedings{hoefler07:_case_for_stand_non_block_collec_operat,
  author =      {Torsten Hoefler and Prabhanjan Kambadur and Richard L. Graham and Galen Shipman and Andrew Lumsdaine},
  title =       {A Case for Standard Non-Blocking Collective Operations},
  booktitle =   {Proceedings, Euro PVM/MPI},
  year =        2007,
  address =     {Paris, France},
  month =       {October}
} 

This Mirror is provided by AmbiWeb GmbH. Contact details can be found in the imprint / Impressum.
The Download-Server is located in central Germany. Servertime is Thursday July 29 12:34:19 EDT 2010