[petsc-users] Parallelizing a matrix-free code

[Praveen C]

> On 16-Oct-2017, at 12:56 PM, Michael Werner <michael.werner at dlr.de> wrote:
> 
> However, since its an unstructured grid, those subgrids are not necessarily made up of points with successive global IDs.

It should be easy to renumber the points so that each partition has contiguously numbered point ids. This is what we do in our CFD code during the stage where we partition the mesh with Metis.

Best
praveen
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20171016/3964e82f/attachment.html>