[MITgcm-devel] MNC "tile assembly" : assemble per-file tiles into a global grid

[Martin Losch]

Hi Ed,

I figured you'd like some feedback on your mnc_assembly.m.
I tried your mnc_assembly script and FAILED (o:
1. I think you should replace all occurances of "&&" and "||" with "&" 
and "|" if you want it to work with Matlab6.1 (don't laugh). After 
making those changes, I could run the test_assem_test.m script 
successfully, that is without obvious errors.
2. In your documentation (help mnc_assembly) your call your second 
argument a "cell" array, but in fact it is a "struct", as can be seen 
from test_assem_test.m (a cell array doesn't work, it took me some time 
to figure that out).
3. I can now run everything on my output (20CPUs, 2D vertical slab, 
much like plume_one_slope), but the resulting all.00000.nc contains 
only the values of one tile/CPU (some tile in the middle). I cannot 
figure out why, so I put now a 140MB gzipped archive on 
sea.mit.edu:/net/sea2/scratch/mlosch/for_ed.tgz for you to grab and see 
what's going, if you find the time. I tried this:
 >> vars=struct([]);
 >> vars(1).name='Temp';
 >>vars(2).name='W';
 >>[n,m]=mnc_assembly('./state.0000.%06d.nc',vnames)
    Found 20 files matching the pattern: "./state.0000.%06d.nc"
Looking ...
Copying ...
n=0
m=0
 >>

Martin

On Dec 10, 2004, at 8:56 PM, Ed Hill wrote:

>
> Hi folks,
>
> A first working version of a MatLAB script that does "MNC tile 
> assembly"
> for both the "exch1" and "exch2" tile geometries is now available.  You
> can get it along with two example data sets at:
>
>  $ cvs co development/edhill/mnc_assembly
>
> and run it using:
>
>  $ cd development/edhill/mnc_assembly
>  $ matlab
>>> mnc_assem_test
>
> There are numerous things that should (and will!) be done to improve 
> the
> script:
>
>  - respect the 2GB limit
>  - allow for selection of arbitrary T and Z levels
>  - better handling of attributes and coord-vars
>
> but this is just a first working (Beta?) version.
>
> 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
>
> _______________________________________________
> MITgcm-devel mailing list
> MITgcm-devel at mitgcm.org
> http://dev.mitgcm.org/mailman/listinfo/mitgcm-devel