[MITgcm-support] Compiling with optfile at NCAR on lightning
Patrick Heimbach
heimbach at MIT.EDU
Mon Jan 23 21:39:56 EST 2006
Hi Galen,
the model has been quite frequently run on
lightning using MPI.
Your problem sounds suspiciously like a setup problem,
as a consequence of which only 1 processor will
compute over part of the domain only.
But hard to tell for sure from your mail.
Maybe send us (off the list) some paths
on lightning where you compile and run
(in case they are world-readable).
-Patrick
On Mon, 2006-01-23 at 19:45, Galen McKinley wrote:
> Has anyone had success compiling on lightning at NCAR with an optfile like
> linux_amd64_pgf77+mpi_ncar?
>
> I am able to run on Lightning if I don't specify an optfile when using
> genmake. However, I need to figure out how to use MPI and NetCDF, so I need
> an optfile.
>
> Using linux_amd64_pgf77+mpi_ncar or similar, I find that the North Atlantic
> setup we are using only reads in the bathymetry up to 70N and then chops
> everything off to the North. All looks fine up to that point. This does not
> happen if I compile without an optfile. I get the same problem with and
> without NetCDF and MPI.
>
> Any suggestions would be most welcome.
>
> Thanks!
> Galen
>
> --------------------------------------------------------------------------
> Galen A. McKinley, Assistant Professor
> Atmospheric and Oceanic Sciences 608 262 4817
> University of Wisconsin - Madison www.aos.wisc.edu/~galen/
> --------------------------------------------------------------------------
>
>
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support
--
--------------------------------------------------------
Patrick Heimbach Massachusetts Institute of Technology
FON: +1/617/253-5259 EAPS, Room 54-1518
FAX: +1/617/253-4464 77 Massachusetts Avenue
mailto:heimbach at mit.edu Cambridge MA 02139
http://www.mit.edu/~heimbach/ USA
More information about the MITgcm-support
mailing list