[petsc-users] Local and Global numberings after DMPlexDistribute

[Anthony Vergottis]

Thanks a lot Matt. That helped.

Regards,
Anthony


On 13 November 2013 00:10, Matthew Knepley <knepley at gmail.com> wrote:

> On Tue, Nov 12, 2013 at 5:32 PM, Anthony Vergottis <a.vergottis at ucl.ac.uk>wrote:
>
>> I would like to calculate the area of all the elements in my unstructured
>> mesh. This is the process that I have envisaged:
>>
>> 1) Partition the mesh with DMPlexDistribute.
>>
>> 2) Obtain which cell/vertices are owned by each process.
>>
>> 3) Have each process calculate the area of the cells it own, as I would
>> like to do this in parallel.
>>
>> I hope this explains things a bit better.
>>
>
> ierr = DMPlexGetHeightStratum(dm, 0, &cStart, &cEnd);CHKERRQ(ierr);
> for (c = cStart; c < cEnd; ++c) {
>   PetscReal vol;
>   ierr = DMPlexComputeCellGeometryFVM(dm, c, &vol, NULL,
> NULL);CHKERRQ(ierr);
>   <do crap with the cell volume>
> }
>
>    Matt
>
>
>> Thanks.
>>
>> Regards,
>> Anthony
>>
>>
>>
>> On 12 November 2013 23:25, Matthew Knepley <knepley at gmail.com> wrote:
>>
>>> On Tue, Nov 12, 2013 at 4:10 PM, Anthony Vergottis <
>>> a.vergottis at ucl.ac.uk> wrote:
>>>
>>>> Dear All,
>>>>
>>>> How does one get the local and global numbers of nodes/elements on each
>>>> process after the DMPlexDistribute function has been used to partition the
>>>> mesh (DM object)?
>>>>
>>>
>>> Can you be more explicit? Vertices and cells are renumbered after
>>> distribution to be contiguous on
>>> each process:
>>>
>>>   cell: [0, numCells)
>>>   vertices: [numCells, numCells+numVertices)
>>>
>>>    Matt
>>>
>>>
>>>> Are there any built in PETSc  functions that offer such functionality?
>>>>
>>>> Thanks in advance.
>>>>
>>>> Regards,
>>>> Anthony
>>>>
>>>
>>>
>>>
>>> --
>>> What most experimenters take for granted before they begin their
>>> experiments is infinitely more interesting than any results to which their
>>> experiments lead.
>>> -- Norbert Wiener
>>>
>>
>>
>
>
> --
> What most experimenters take for granted before they begin their
> experiments is infinitely more interesting than any results to which their
> experiments lead.
> -- Norbert Wiener
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20131113/5052b7cb/attachment.html>