[MITgcm-support] Re: mpi on gfdl origin
Uli Riemenschneider
uriemenschneider at whoi.edu
Wed Sep 1 12:52:23 EDT 2004
Thanks for the reply Tom (and others!),
I got the model running fine now, I think. Am waiting for some output!
U.
Tom Haine wrote:
> I'm running mitgcm with mpi at GFDL with no problems (now!). I use
> globalFiles=.TRUE. and it works OK (although it doesn't work on my
> opteron cluster). I haven't experimented with useSingleCpuIO=.true. yet.
>
> I have no problem with job control and receive all the usual STD output.
> My job script is:
>
> # run on 81 cpus for 480min
> #$ -pe lsc.alloc 81 -l h_cpu=8:00:00
> #$ -N mitgcmuv -o /home/twh/temp/run1.out
> cd $FTMPDIR
> echo $FTMPDIR
> cp ~/MITgcm/twnh/spill_jet_2km/run1/bin/mitgcmuv .
> cp ~/MITgcm/twnh/spill_jet_2km/run1/input/* .
> cp /archive/twh/forcing/run1/* .
> cp /archive/twh/spill_jet_2km/results/run1/pickup*21600* .
> ls -alt
> mpirun -np 81 mitgcmuv
> ls -alt
> # Take care to write data in empty directory!!!!!
> /bin/rm -r /archive/twh/spill_jet_2km/results/out
> mkdir /archive/twh/spill_jet_2km/results/out
> mv * /archive/twh/spill_jet_2km/results/out
>
> Tom.
> --------------------------------------------
> Thomas W. N. Haine
>
> Associate Professor of Physical Oceanography,
> Department of Earth & Planetary Sciences,
> 329 Olin Hall, The Johns Hopkins University,
> Baltimore, MD 21218, USA.
> Tel : 410 516 7048, Fax : 410 516 7933
> Thomas.Haine at jhu.edu
> http://www.jhu.edu/~eps/faculty/haine
> --------------------------------------------
>
>
--
*********************************************************
Ulrike Riemenschneider, Postdoctoral Investigator
Physical Oceanography Dept. MS #21
Woods Hole Oceanographic Institution
Woods Hole, MA 02543, USA
Phone: (+1) 508 289 2916 Fax: (+1) 508 457 2181
*********************************************************
More information about the MITgcm-support
mailing list