Could an alternartive be to read it as a serial triangulation and then to partition it as a fullydistributed triangulation using METIS or ZOLTAN? Sure you would loose the ability to do further dynamic mesh adaptation, but still could be useful
On Wednesday, July 16, 2025 at 7:28:58 a.m. UTC+2 [email protected] wrote: > Actually, I think there is enough information to recreate the full > triangulation, provided that you save *one file per processor*, and read > back one file per processor. > > The reason why this should work, is because the > “serialize_triangulation=true”, actually serialises the full triangulation > using the serialization mechanisms, and stores that information in a field > of the (vtu) xml file that is ignored by paraview. However, the > triangulation you have is effectively different on each processor, and you > must make sure that you read it back in on each processor in the correct > way. > > L. > > > On 16 Jul 2025, at 05:32, 'Wolfgang Bangerth' via deal.II User Group < > [email protected]> wrote: > > > > On 7/15/25 07:16, LY XXXiao wrote: > >> What we want: > >> There is a distributed refined mesh1 with defined parameters. This mesh > has been saved as *.vtu file, with 'serialize_triangulation=true'. > >> And we want to read it in from the vtu file as a distributed > triangulation. > >> What we have done: > >> As you suggested, we attached the distributed triangulation to the > gridin and used read_vtu(), the n_global_active_cells() is however, 0. I > have attached the test snippet here. Hopefully, you or someone else will > have the time to correct my understanding of this function. > > > > I don't think you can read in distributed, refined triangulations that > way. (Not in the sense of "this isn't implemented", but "this is > conceptually not possible".) If you want to write and read distributed, > refined triangulations, you should use the serialization mechanism for > triangulations. VTU is not the right format for this, there is not enough > semantic information in VTU files to re-create triangulations that way. > > > > Best > > W. > > > > -- > > ------------------------------------------------------------------------ > > Wolfgang Bangerth email: [email protected] > > www: http://www.math.colostate.edu/~bangerth/ > > > > > > -- > > 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 [email protected]. > > To view this discussion visit > https://groups.google.com/d/msgid/dealii/23459386-0c6a-48b7-82a6-b27df88acff2%40colostate.edu > . > > -- 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 [email protected]. To view this discussion visit https://groups.google.com/d/msgid/dealii/71b1bbac-18b7-4730-9af5-d65c87bd4600n%40googlegroups.com.
