[Getdp] getdp + petsc in parallel - partitioning ?

Christophe Geuzaine cag32 at case.edu
Thu Apr 20 14:47:42 CEST 2006


Thomas Jung wrote:
> Hello,
> 
> I have built getdp with petsc- and metis support,
> trying to run it in parallel, I get this:
> 
>      Error     : 1 partitions on 2 CPU
> 
> Seems logical, I probably have to partition the mesh (or the matrix ????)
> somehow using metis, but how do I do that ?
> 

The parallel PETSc version currently only works on specific test 
problems, and will give *wrong results* in general. Please don't use it!

You can safely use the single processor PETSc version, though: it has 
been extensively tested both in real and complex arithmetic.

Christophe

> Again, thanks for any hint !
> 
>  Thomas Jung
> _______________________________________________
> getdp mailing list
> getdp at geuz.org
> http://www.geuz.org/mailman/listinfo/getdp
> 


-- 
Christophe Geuzaine
Assistant Professor, Case Western Reserve University, Mathematics
http://www.case.edu/artsci/math/geuzaine