Newsgroups: comp.parallel.mpi
From: torj@zapffe.mat-stat.uit.no (Tor Johansen)
Subject: Re: Problem using LAM5.2
Organization: University of Tromsoe, Norway
Date: 1 Dec 1995 09:44:25 GMT
Message-ID: <49mipp$hvv@news.uit.no>

In article <49m28t$6o4@nuscc.nus.sg>, wongwoon@iscs.nus.sg (WWF) writes:
|> 
|> Hi,
|> 	Currently, i have a piece of MPI code and it runs successfully 
|> using MPICH.  However, with the same piece of code, it hangs half way 
|> through the execution when using LAM5.2.  May i know what could be the 
|> possible reasons?  I'm running the processes on a SP2.  Thanx in advance!

Hi,
maybe the default buffer-size used by MPI? I used MPI_SENDRECV_REPLACE
on a Paragon, and it work well. Moving to the SP2 this caused the program
to the program hang... Replacing MPI_SENDRECV_REPLACE by MPI_IRECV and 
MPI_SEND and explicit introducing an extra vector (used in MPI_IRECV)
in my program removed this problem. This was maybe not an elegant way to
solve my problem - but it worked fine.

Sincerely

Tor Johansen

   __________________________________________________________________
  |                                                                  |
   \              INSTITUTE OF MATHEMATICAL SCIENCES                 |
  o/\_                      Dept. of Physics                        _/\o
 <\__,\        University of Troms, N-9037 Troms, Norway,        |,__/>
  ">   |         Tlf. +47 77 64 40 00, Fax +47 77 64 47 65         \  <"
   `   |    Tor Johansen, Ph. D. st., direct call +47 77 64 52 08,  \ '
        \                  E-mail: thorj@mack.uit.no                |
         \__________________________________________________________|

