[Gmsh] partition issue version 2 :-))

Umut Tabak u.tabak at tudelft.nl
Tue Jan 18 04:37:46 CET 2011


Christophe Geuzaine wrote:
> On 14/01/11 01:46, Umut Tabak wrote:
>
> Hi Umut,
>
> I'm not quite sure I understand your question. The third tag tells how 
> many partitions an element belongs to; the next tags give these 
> partitions' ids. If a partition id is negative, it denotes a partition 
> in which the element is a ghost cell.
>
> Is the confusion linked to the negative ids? (At the moment there is 
> no option to disable ghost cells--you need to recompile from the 
> sources.)
Dear Cristophe,

The confusion is related to the two attached output files, one is a 2d 
mesh and the other is a 3d mesh. In the 2d case, 3rd entry of the tags 
directly give the partition number where there are no ghost cells.

If the explanation that you provided works than the 2d mesh definition 
is wrong but the second is right, I mean box333_2parts.msh. Because in 
that file, exactly the 3rd entry gives how many partitions an element 
belongs to and ids follow accordingly but in the 2d file that is not the 
case.

This point confused me.

Best regards,
Umut


-------------- next part --------------
A non-text attachment was scrubbed...
Name: parts2Unsym.msh
Type: model/mesh
Size: 278597 bytes
Desc: not available
URL: <http://www.geuz.org/pipermail/gmsh/attachments/20110118/f4449548/attachment.msh>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: box333_2parts.msh
Type: model/mesh
Size: 2945 bytes
Desc: not available
URL: <http://www.geuz.org/pipermail/gmsh/attachments/20110118/f4449548/attachment-0001.msh>