[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: Replacing LAM with OpenMPI in Fedora Core



On Mon, 2006-01-02 at 12:26 -0500, Deji Akingunola wrote:
> >From the frontpage of the lam's website (http://www.lam-mpi.org/);
> "LAM/MPI is now in a maintenance mode. Bug fixes and critical patches
> are still being applied, but little real "new" work is happening in
> LAM/MPI. This is a direct result of the LAM/MPI Team spending the vast
> majority of their time working on our next-generation MPI
> implementation -- Open MPI."
> The site also states that, "Since it's an MPI implementation, you
> should be able to simply recompile and re-link your applications to
> Open MPI -- they should "just work." Open MPI contains many features
> and performance enhancements that are not available in LAM/MPI."

Hi Deji and everyone else interested in MPI,

I understand that OpenMPI is supposed to [at least, for some people's
perception! ;-)] become the "one true MPI" implementation that eclipses
all others due to its very cool new modular design and other features.
But, even so, I think it would be a good idea to have and maintain other
MPI versions (such as MPICH v1 & v2, LAM, etc.) in Extras so that people
have some flexibility.  And to do that, we'll very likely need to setup
the multiple MPI packages using alternatives.

The OpenMPI and MPICH2 packages that we (Deji and I) have submitted to
FE:

  https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=173719
  https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=171993

are not yet setup for alternates but that certainly can be done -- it
probably wouldn't be too hard to add that capability.

So, my questions are:

 1) Who are the LAM maintainer(s) within Red Hat and can they 
    *please* step forward and discuss the Fedora plans for MPI
    in general and LAM in particular?

 2) If its possible, could we please modify LAM so that it can 
    work with other MPI implementations through alternatives?

 3) And could we perhaps move LAM out of Core?  Or maybe 
    substitute OpenMPI for LAM if one MPI must remain as the 
    default in Core?

I know that this will take some work but we've already made some
progress and now its time for someone within Red Hat to speak up and
work with us.  Please!

Ed

-- 
Edward H. Hill III, PhD
office:  MIT Dept. of EAPS;  Rm 54-1424;  77 Massachusetts Ave.
             Cambridge, MA 02139-4307
emails:  eh3 mit edu                ed eh3 com
URLs:    http://web.mit.edu/eh3/    http://eh3.com/
phone:   617-253-0098
fax:     617-253-4464


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]