[Dune] parallel coarsening

[Marco Cisternino]

Hi Andreas,
I know, I would like to provide an example but I never experienced such 
a situation in a simple case.
I'm studying an hypersonic 2D case in 3D and I notice this behaviour 
because I loose the geometrical z-uniformity after the third call of the 
adaptation function. It happens only with coarsening. If I exclude 
coarsening, making only refinement, everything works in the right way. I 
think that's because refinement happens anyway, no matter how neighbours 
are marked.
Could you sketch the right way to proceed in marking elements in 
parallel alugrid, please? As synthetic as you can.
However, I'll try to store marks in a container, update its ghost and 
the mark elements. Could this be an efficient way?
Thanks a lot,

Marco



Il 11/06/2013 11:34, Dedner, Andreas ha scritto:
> Hi.
> Would be good to have a simple test case with as few elements as possible.
> Could you provide something?
> Best
> Andreas
> ________________________________________
> From: dune-bounces+a.s.dedner=warwick.ac.uk at dune-project.org [dune-bounces+a.s.dedner=warwick.ac.uk at dune-project.org] on behalf of Marco Cisternino [marco.cisternino at optimad.it]
> Sent: 11 June 2013 10:21
> To: dune at dune-project.org
> Subject: [Dune] parallel coarsening
>
> Hi duners,
> I have an issue about parallel coarsening.
> In order to mark elements for coarsening in alugrid, I loop over
> interiorBorder elements, check for the indicator and mark elements below
> a certain threshold. However, even if some border elements are marked
> for coarsening, they are not coarsened without any logical reason. The
> only thing I can imagine is that ghost have not been treated. What do
> you suggest?
> Is it better to store marks in a container, update ghosts, then mark all
> and adapt?
> Thank you.
>
> Marco
>
>
> --
> ------------------------------------
> Marco Cisternino, Ph.D.
> OPTIMAD Engineering s.r.l.
> Via Giacinto Collegno 18
> 10143 Torino - Italy
> www.optimad.it
> marco.cisternino at optimad.it
> +39 011 19719782
> ------------------------------------
>
>
> _______________________________________________
> Dune mailing list
> Dune at dune-project.org
> http://lists.dune-project.org/mailman/listinfo/dune
>
>

-- 
------------------------------------
Marco Cisternino, Ph.D.
OPTIMAD Engineering s.r.l.
Via Giacinto Collegno 18
10143 Torino - Italy
www.optimad.it
marco.cisternino at optimad.it
+39 011 19719782
------------------------------------