[MITgcm-devel] [MITgcm-cvs] MITgcm/eesupp/src CVS Commit

[Matthew Mazloff]

Hi Dimitris

This won't really work as I think the problem is that the processors are slightly out of sync...
Letting each processor write its own scratch file is a much safer way. And its already coded -- just define TARGET_BGL or TARGET_CRAYXT

I set 
#define TARGET_CRAYXT
in CPP_EEOPTIONS.h

-Matt



On Sep 29, 2013, at 8:06 AM, Dimitris Menemenlis <dimitri at forge.csail.mit.edu> wrote:

> Update of /u/gcmpack/MITgcm/eesupp/src
> In directory forge:/tmp/cvs-serv16055
> 
> Modified Files:
> 	open_copy_data_file.F 
> Log Message:
> As processor count was increased on pleaides, there was an occasional race
> condition on lustre files system that caused a process to attempt to read
> scratch1* (parameter file without comments) before it was avaialble. As a
> temporary, probably not fail-safe fix, I have added: CALL SYSTEM('sleep 1')
> 
> 
> _______________________________________________
> MITgcm-cvs mailing list
> MITgcm-cvs at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-cvs