Abbas, IIRC it's not allowed to refine the mesh after it has been distorted since the new vertex locations won't communicated correctly when the mesh is redistributed.
Best, Daniel On Wed, Mar 19, 2025 at 1:02 PM Abbas Ballout <abbas.ballou...@gmail.com> wrote: > > The mesh is generated then distorted. > The code after that enters a cycle, assemble, solve, refine -> enter cycle, > assemble, solve, refine -> enter cycle .... > I just slapped a distort random right after mesh generation in step 40 and > it's creating that artifact after a couple of runs with more than 1 MPI core. > Seemed like a bug to me, but I was hoping someone would replicate it just to > be sure. > > > On Mon, Mar 17, 2025, 13:17 blais...@gmail.com <blais.br...@gmail.com> wrote: >> >> That's a very good question Abbas, I am actually not sure. >> If I understand correctly, you made a distributed parallel triangulation, >> then distorted it then you refined it is that it? Or did you distort it >> after refining it? >> >> I'll look at the code you provided and try to identify what's going on :) >> >> >> On Tuesday, February 25, 2025 at 8:41:29 a.m. UTC-5 abbas.b...@gmail.com >> wrote: >>> >>> Hello, >>> >>> Taking step 40 and using the GridTools::distort_random function, the mesh >>> seems to look "broken" after a couple of AMRs when running with more than 1 >>> mpi process (see attached screenshot ran with mpi 4). Is this normal ? This >>> doesn't look like this is an expected behaviour after looking at the docs. >>> The only change I made was to add the GridTools::distort_random function >>> after the grid generation. (step-40.cc file is attached bellow) >>> >>> Using shared::tria doesn't seem to suffer from the same issue but it breaks >>> saying: >>> "A parallel::shared::Triangulation needs to be refined in the same way >>> on all processors, but the participating processors don't agree on the >>> number of active cells." >>> >>> Best, >>> Abbas >> >> -- >> The deal.II project is located at http://www.dealii.org/ >> For mailing list/forum options, see >> https://groups.google.com/d/forum/dealii?hl=en >> --- >> You received this message because you are subscribed to a topic in the >> Google Groups "deal.II User Group" group. >> To unsubscribe from this topic, visit >> https://groups.google.com/d/topic/dealii/U3QfnIvWq0I/unsubscribe. >> To unsubscribe from this group and all its topics, send an email to >> dealii+unsubscr...@googlegroups.com. >> To view this discussion visit >> https://groups.google.com/d/msgid/dealii/bfff9e21-e7e8-46f6-aeb1-a1cd4c4dcdc1n%40googlegroups.com. > > -- > The deal.II project is located at http://www.dealii.org/ > For mailing list/forum options, see > https://groups.google.com/d/forum/dealii?hl=en > --- > You received this message because you are subscribed to the Google Groups > "deal.II User Group" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to dealii+unsubscr...@googlegroups.com. > To view this discussion visit > https://groups.google.com/d/msgid/dealii/CAHZ_B5diz7NKQzb2ajkRVF019m5mfGQfPfOQw8SJ_j2P%3D2tLRA%40mail.gmail.com. -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. To view this discussion visit https://groups.google.com/d/msgid/dealii/CAOYDWbJxh4fB_dJKyA3Jx91mEWFPjLFjGjL0hH8gK%3DrHRjnUoA%40mail.gmail.com.
