From: Jay Walter Larson <Jay.Larson@anu.edu.au>
Message-Id: <199510230016.KAA27223@leonard.anu.edu.au>
Subject: Re: hardware for ccm2
To: gdc3491@csrp.tamu.edu (Gordon Carrie)
Date: Mon, 23 Oct 1995 10:16:38 +1000 (EST)
In-Reply-To: <v01510101acadd03bfb5a@[128.194.71.249]> from "Gordon Carrie" at Oct 20, 95 04:22:36 pm
Hi Gordon,
>
> This is in response to James Broermann's posting.
>
> I run CCM2/T42 on a Cray J90, where it takes about 600 CPU seconds / model
> day. Parellelizing on 2 processors increases speed by about 40%.
>
> I have tried to run it on SGI Power Challenge, but could never get it to
> parallelize effectively. I would be interested to know if anyone else has
> tried.
>
> Regards,
> Gordon
>
> ---------|---------|---------|---------|---------|---------|---------|--
>
> Gordon D. Carrie
> Graduate Student of Dr. Ken Bowman
> at
> Climate Systems Research Project
> Department of Meteorology
> Texas A&M University
> College Station, Texas 77843-3150
>
> Oceanography and Meteorology Building 1013, 409-845-0189
>
> Home:
> 409-779-2228
> Ejfx@aol.com
If you are in search of a parallel version of CCM2, the USA DOE Computer
Hardware, Advanced Mathematics, and Model Physics (CHAMMP) program has a
project whose aim is just that. You might want to check out this URL:
http://www.mcs.anl.gov/home/itf/climate/index.html
I believe the official release of PCCM2 has occurred, or will in the
near future. I'm not quite sure if it's up for anonymous ftp yet...
Good Luck,
Jay