[MITgcm-support] Optimizing the 2D pressure solver

[Matthew Mazloff]

Hi Christopher,

The issue may be that a  1792 x 448 x 20 domain may not efficiently  
scale up to 1000 processors; probably more like 800.  The overlap to  
interior cell ratio is getting too large.

-Matt


On Sep 21, 2007, at 7:04 PM, Dimitris Menemenlis wrote:

> Christopher, when we ran on 4000 processors for a global 1/16th  
> configuration, we used:
>
>  cg2dMaxIters=100,
>  cg2dTargetResidual=1.E-5,
>
> D.
>
> -- 
> Dimitris Menemenlis <menemenlis at sbcglobal.net>
> 5056 Oakwood Ave, La Canada, CA 91011-2450
> tel/fax: 818-790-6735;   cell: 818-625-6498
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support