Newsgroups: comp.parallel.mpi
From: sungeun@cs.washington.edu (Sung-Eun Choi)
Subject: SGI's native MPI
Keywords: SGI
Organization: Computer Science & Engineering, U of Washington, Seattle
Date: Mon, 26 Aug 1996 18:13:28 GMT
Message-ID: <DwrByG.uo@beaver.cs.washington.edu>

Hello,

I'm new to MPI, so if this has been discussed already, just point me
to the archive.

I have a question about SGI's native implementation of MPI for the
PowerChallenge.  If you have experience with this, please read on..

I've ported a "working" MPI application ("working" under the MPICH
implementation) to the PowerChallenge.  In doing so, I rewrote the
code to abide by the restrictions of the SGI implementation as stated
in the man page.  Specifically, by default, SGI's MPI_Send() does not
buffer shared memory messages that are longer than 80 bytes in length.

They suggested two ways around the problem:

     1) Rewrite the code to use either MPI_Isend() or MPI_Bsend().

     2) Set MPI_SHMEM_BUFFER_THRESHOLD as needed.

I chose the first option, but it did not solve the problem.  I still
need to set MPI_SHMEM_BUFFER_THRESHOLD (which incurs a performance
hit).

Does anyone know if there are other restrictions that are not
documented in the man pages?  Is there any on-line documentation other
than the man pages?

Thanks a ton for any help!

-- Sung

