Open MPI logo

Open MPI User's Mailing List Archives
  |   Home   |   Support   |   FAQ   |   all Open MPI User's mailing list
 About
  Publications
  Open MPI Team
  FAQ
  Videos
 Open MPI Software
  Download
  Documentation
  Source Code Access
  Bug Tracking
  Regression Testing
  Version Information
 Sub-Projects
  Hardware Locality
  MPI Testing Tool
  Portable Linux Processor Affinity
  Open MPI User Docs
  Open Tool for Parameter Optimization
 Community
  Mailing Lists
    OMPI announce archives
    OMPI users archives
    OMPI devel archives
    OMPI docs archives
    hwloc announce archives
    hwloc users archives
    hwloc devel archives
    MTT announce archives
    MTT users archives
    MTT developers archives
    ORCM users archives
    ORCM devel archives
    PLPA announce archives
    PLPA users archives
  Getting Help/Support
  Contribute
  Mirrors
  Contact
  License
 
Date view Thread view Subject view Author view

Subject: Re: [OMPI users] Progress of the asynchronous messages
From: Eugene Loh (Eugene.Loh_at_[hidden])
Date: 2008-11-06 12:52:28

vladimir marjanovic wrote:
In order to overlap communication and computation I don't want to use MPI_Wait.
Right.  One thing to keep in mind is that there are two ways of overlapping communication and computation.  One is you start a send (MPI_Isend), you do a bunch of computation while the message is being sent, and then after the message has been sent you call MPI_Wait just to clean up.  This assumes that the MPI implementation can send a message while control of the program has been returned to you.  The experts can give you the fine print, but my simple assertion is, "This doesn't usually happen."

Rather, the MPI implementation typically will send data only when your code is in some MPI call.  That's why you have to call MPI_Test periodically... or some other MPI function.
For sure the message is being decomposed into chucks and the size of chuck is probably defined by environment variable.
Maybe do you know how can I control size of chuck?
I don't.  Try running "ompi_info -a" and looking through the parameters.  For the shared-memory BTL, it's mca_btl_sm_max_frag_size.  I also see something like coll_sm_fragment_size.  Maybe look at the parameters that have "btl_openib_max" in their names.
Date view Thread view Subject view Author view
[Flag]
 Indiana University (Main Open MPI server)
This site is located in: Bloomington, IN, USA
Page last modified: 6-Nov-2008
©2004-2012 The Open MPI Project