[Carpet] Changes to regridding

Bela Szilagyi szilagyi at aei.mpg.de
Thu May 4 16:19:24 CEST 2006


For a few days now I am systematically being bit by this issue.  I have a 
simulation where, on the fly, I am dropping the finest level.

The recovery, using the original parfile, fails.  The "solution" is to edit 
the parfile and lower the parameter value for  refinement levels.

Is there a better solution than having to edit the parfile between checkpoint 
and recovery?



On Monday 01 May 2006 14:14, Erik Schnetter wrote:
> > Not mentioning that it is possible to get core dumps by trying to  
> > change the
> > grid-structure at recovery,
>
> Hm.  You are not really supposed to do that.  Carpet has currently a  
> missing feature, namely that it does not store the grid structure in  
> the checkpoint file.  It is therefore necessary to specify the  
> correct grid structure in the parameter file that you use to  
> recover.  It is, within certain bounds, possible to change the grid  
> structure (e.g. having fewer levels), but that was not originally  
> intended and is not really supported.  It is also not strictly  
> necessary, since there will be a regridding step after recovering and  
> before the next time step.

-- 
Bela Szilagyi
----------------------------------------------------
Max-Planck-Institut für Gravitationsphysik
Albert-Einstein-Institut
Tel: +49 331 567 7632
Fax: +49 331 567 7649
----------------------------------------------------




More information about the developers mailing list