[MITgcm-devel] FIZHI on Columbia

[Ed Hill]

Hi Andrea, Jean-Michel, and Chris,

I have a 2-CPU MPI version of MITgcm w/ FIZHI ("fizhi-cs-32x32x10")
running on Columbia and you can see the results at:

http://mitgcm.org/testing.html
http://mitgcm.org/testing/results/2005_08/tr_Columbia-fizhi-new_20050808_1/summary.txt

All of the testreport quantities have 3--6 digits of agreement and are
similar in magnitude to the other 2-CPU MPI runs of "fizhi-cs-32x32x10"
such as:

http://mitgcm.org/testing/results/2005_08/tr_cg01_pgi_20050807_0/summary.txt
http://mitgcm.org/testing/results/2005_08/tr_itrda-pgf77_20050808_0/summary.txt

So, according to Jean-Michel, its very likely that all the input fields,
parameters, etc. are being read correctly.

The changes that I made to get it working are:

 1) Increase maximum stack size:  ulimit -s unlimited

 2) The model kept dying with end-of-file errors when trying to 
    read "veg19232.data" so I wrote an alternate and optional 
    version that allows the veg-tile data to be read, written, 
    and even on-the-fly-converted to netCDF format.

I would like to check the code modifications from (2) into the model but
will wait to discuss it with the rest of you -- especially Andrea.

Ed

-- 
Edward H. Hill III, PhD
office:  MIT Dept. of EAPS;  Rm 54-1424;  77 Massachusetts Ave.
             Cambridge, MA 02139-4307
emails:  eh3 at mit.edu                ed at eh3.com
URLs:    http://web.mit.edu/eh3/    http://eh3.com/
phone:   617-253-0098
fax:     617-253-4464