[MITgcm-support] file size issue with mpi

Uli Riemenschneider uriemenschneider at whoi.edu
Tue Aug 31 12:59:20 EDT 2004


Hi,

can I follow up on the file size issue ? I have started running (or 
well, trying to run) the MITgcm on the supercomputer in Princeton (SGI 
Origin) and discovered that the output files are not readable. I wrote 
out global files, but I gather with 25 processors that's a bad idea! 
Will "useSingleCpuIO" solve the problem, or should I be writing out 
tiled files?

Also, my run produced no STDOUT.*.* files, but instead printed the 
monitored fields for all processors to the screen in one big muddled 
heap. This never happened before with previous multiprocessor runs on 
our local SGI. Is there a flag which needs to be set to make sure the 
STD... files are written properly?

Thanks,
Uli

Dimitris Menemenlis wrote:
> Tom and Michael, the "useSingleCpuIO" output option creates a single
> output file per field, similar to the "globalFiles" option.  But it
> is a lot more robust and, at least on the SGI platforms (Origin and
> Altix), many times faster.
> 
> Dimitris
> 
> --
> Dimitris Menemenlis <menemenlis at jpl.nasa.gov>
> Jet Propulsion Lab, MS 300-323
> 4800 Oak Grove Dr., Pasadena CA 91109
> tel:818-354-1656    fax:818-393-6720
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://dev.mitgcm.org/mailman/listinfo/mitgcm-support
> 
> 


-- 
*********************************************************
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