From mboxrd@z Thu Jan  1 00:00:00 1970
Message-ID: <1fb06423f1ac109c8ba323c10a5f3b70@iram.fr>
To: 9fans@9fans.net
From: "Mathieu Lonjaret" <mathieu.lonjaret@gmail.com>
Date: Thu, 16 Sep 2010 11:37:26 +0200
MIME-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Subject: [9fans] [OT] CSP in fortran ?
Topicbox-Message-UUID: 577bcf7e-ead6-11e9-9d60-3106f5b1d025

Hi all,

A bit offtopic, but since I know some HPC and/or csp pros are reading
9fans...

I have to work on a piece of fortran code to break it into pieces that
would run concurrently (=C3=A0 priori SPMD kind) , typically on a multico=
re
machine, not really on a big HPC cluster. One of the requirements is
to bring the least amount of modifications to the current code (for
maintainability).

I was hoping to find a fortran lib similar to p9p libthread, but not
much luck so far. Does anyone know of something like that?
I have started tackling it with CoArrays (http://www.co-array.org/),
and I'm going to try with MPI next. What would you guys recommend between
the two for that kind of problem?

Thanks,
Mathieu