[MITgcm-support] cpus on land points

[Jody Klymak]

Hi all,

On Jul 26, 2013, at  17:14 PM, Gianmaria Sannino <gianmaria.sannino at enea.it> wrote:
> I was wondering if there is any way for MITgcm to avoid using cpus running only land points?


I've wondered this too - i.e. if there was an easy mechanism to allow the user to specify which tiles neighbour which other ones.  It would help a lot in designing grids for fjords or other oddly-shaped bodies of water.  

Bonus points would be for allowing each grid to specify its own dx and dy independently of other grids (except for the adjacent neighbour, of course).  You could imagine convoluted structures:

G9  G8  G7
G10     G6
G11     G5
        G4
G1  G2  G3

One could imagine if G10 and G11 were your domain of interest, and G1-9 somewhere with larger physical extent that you might want G10 and G11 to have smaller dy than G6 and G5, and smaller dx than G1 (though of course the same dx as G9).  
  
If not already implemented, pointers on how to hack this into the code would be very welcome.  


Thanks,   Jody


> Ciao
> Gianmaria
> 
> 
> 
> 
> 
> MESSAGGIO ISTITUZIONALE
> INSIEME PER UNA GRANDE IMPRESA
> 
> Destina il 5 per 1000 all'ENEA per ricostruire la Città della Scienza 
> Utilizzeremo il tuo contributo e le nostre competenze tecnico scientifiche per progetti di divulgazione scientifica con Città della Scienza. 
> Il nostro codice fiscale è 01320740580
> 
> Per maggiori informazioni: www.enea.it _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support
> 

--
Jody Klymak    
http://web.uvic.ca/~jklymak/




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mitgcm.org/pipermail/mitgcm-support/attachments/20130726/aaae63d4/attachment.htm>