[MITgcm-support] Problems with mpi on beowulf

samar khatiwala spk at ldeo.columbia.edu
Sat Sep 18 09:43:01 EDT 2004


Hi Yuan

I run the model on a Beowulf with mpich (gm) and pgf77. Works fine.

> >Warning: no access to tty (Bad file descriptor).
> >Thus no job control in this shell.

Ignore this. pbs always seems to generate this.

>
> This error message shows that the code can't read eedata, (the default
> folds was changed to my home directory instead of workdir!!)I defined
> "usingMPI = .TRUE." and moved all programs under workdir to home
> directory, then code will run, however the results of simulation are
> obviously wrong comparing to single cpu. Here are the questions:

In your pbs script you need:
cd $PBS_O_WORKDIR

I assume the nodes are NFS mounted so they can all see your working
directory.

Samar



More information about the MITgcm-support mailing list