Rich:

Now that both of our GCM's (TIEGCM and TIME-GCM) have been ported
(Cray, IBM, SGI, Compaq) and parallelized (shared and/or distributed 
memory with 1-d decomposition over latitude), it is time to redesign
and modernize the data structure with the potential for higher horizontal
resolution, and a 2-d data decomposition over the horizontal domain.

I have done some thinking about this, and would like to meet with you
before I dive in and try to fly something that will crash in the end.
(that way, if it does crash, I won't have to take the blame alone ;-))

I have taken the smaller of the 2 models (TIEGCM), and stripped out
the "model layer", leaving a skeletal code that does everything
(user input, initialization, runs parallel, does netcdf history i/o, 
etc) *except* the science.  If we can sketch out a data structure 
strategy, I can experiment with this skeleton model as a prototype
by gradually reintroducing rewritten model layer routines.

Could we meet sometime this month (other than 7/16-18)?  I will be 
gone the first 3 weeks of August, so if not in July, it would have to be 
late Aug or Sept.  I am cc'ing Cecelia, as I have talked w/ her some 
about this, she is somewhat familiar with our codes, and might be willing 
to contribute, esp wrt the software frameworks effort.

--Ben

-----------------------------------------------------------------------
Ben Foster		      	High Altitude Observatory (HAO)
foster@.ucar.edu		phone: 303-497-1595  fax: 303-497-1589  
Nat. Center for Atmos. Res.     P.O. Box 3000 Boulder CO 80307 USA
-----------------------------------------------------------------------