Newsgroups: comp.parallel.pvm
Path: ukc!uknet!EU.net!Germany.EU.net!netmbx.de!zib-berlin.de!uni-paderborn.de!urmel.informatik.rwth-aachen.de!newsserver.rrzn.uni-hannover.de!solaris.rz.tu-clausthal.de!pce60.rz.tu-clausthal.de!stub
From: stub@pce60.rz.tu-clausthal.de (Ulf Bartelt)
Subject: Re: PVM3.2.3 on DEC-Fortran 3.2
Message-ID: <1994Apr12.075952.17764@solaris.rz.tu-clausthal.de>
Sender: root@solaris.rz.tu-clausthal.de (Operator)
Organization: Techn. Univ. Clausthal
X-Newsreader: TIN [version 1.2 RZ-TUC PL2]
References: <1994Apr11.145808.2293@solaris.rz.tu-clausthal.de>
Date: Tue, 12 Apr 1994 07:59:52 GMT
Lines: 32

Me again... ;-)

Ulf Bartelt (stub@pce60.rz.tu-clausthal.de) wrote:
: I'm trying to run benchmarks with PVM on Ultrix (PMAX) systems using
: Fortran77.

: There are massive problems. Somtimes a parallel application crashes
: the system down into the boot monitor, sometimes the same application
: runs correctly...

Meanwhile I tried the same sources under Fortran 3.3 on OSF/1 and the
three programs I tested ran ok as far as I can say after a few hours
of experimenting on 4 ALPHAs...

Having the code dePARMACSed under Linux and Ultrix, the Makefiles had
the hooks for a third system, so testing under OSF/1 was mainly changing
the Makefile and recompiling.

This gives me enough reasons to think, the crashes are a problem of
the PVM/Fortran/ULTRIX combination as the same code has no problems
under (a single CPU) Linux-PVM and four DEC-Alphas...

Maybe even Ultrix isnt the problem and I only have misconfigured some
Ultrix specific part of PVM but if so, I donnow what or where...

Bye for now...

   Ulf.

--
Ulf Bartelt | 2:2437/120.666@fidonet.org | stub@zaphod.in.tu-clausthal.de
            |                            | stub@linux.rz.tu-clausthal.de

