[petsc-dev] DMPlex overlap redistribution

[Matthew Knepley]

On Sun, Mar 3, 2024 at 9:18 PM Adrian Croucher <a.croucher at auckland.ac.nz>
wrote:

> On 4/03/24 2:36 pm, Adrian Croucher wrote:
> > I just re-tested my code with this "no cells in the interface" check
> > commented out of the PETSc code, and it appears to run fine.
>

I put in that check to look for improper handling of overlap cells. The
idea here is that partitions
should be non-overlapping, and then we recreate the overlap after
distribution, which is the easy
way to get parallel consistency.


> Whoops, no it doesn't... I checked more closely and there is another error:
>
> [0]PETSC ERROR: Petsc has generated inconsistent data
> [0]PETSC ERROR: New depth 2 range [0,141) overlaps with depth 1 range
> [0,141)
>

So here I am guessing that the interior cells (rock matrix) mess with the
stratum consistency
(ordering of dimension by stratum in the DAG). So some questions:

1) Are rock matrix cells always associated with only one regular cell?

2) Do the rock matrix cells hold values that need to be communicated?

3) Do you use the topological operations for them, or might we reproduce
this effect by just
    marking the cells with rock matrix cells in them using a DMLabel?

  Thanks,

     Matt


> [0]PETSC ERROR: #1 DMPlexCreateDepthStratum() at
> /home/acro018/software/PETSc/code/src/dm/impls/plex/plex.c:4087
>
> So maybe something else in my code has been broken by recent changes.
> It's a bit hard to come up with a minimal example because for these
> dual-porosity meshes it is doing some fairly complex low-level surgery
> on the DMPlex, to add the "rock matrix" cells "inside" the existing cells.
>
> - Adrian
>
> --
> Dr Adrian Croucher
> Senior Research Fellow
> Department of Engineering Science
> Waipapa Taumata Rau / University of Auckland, New Zealand
> email: a.croucher at auckland.ac.nz
> tel: +64 (0)9 923 4611
>
>

-- 
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

https://urldefense.us/v3/__https://www.cse.buffalo.edu/*knepley/__;fg!!G_uCfscf7eWS!ccaTGf6hASLD8J_yyrdCI75x9u7qNlqDEUgf2_zqm1eb093yrhtvF3azkLw9Ev_djGec-Q9eR7oM2zdqJ1y0$  <https://urldefense.us/v3/__http://www.cse.buffalo.edu/*knepley/__;fg!!G_uCfscf7eWS!ccaTGf6hASLD8J_yyrdCI75x9u7qNlqDEUgf2_zqm1eb093yrhtvF3azkLw9Ev_djGec-Q9eR7oM24SolK9J$ >
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20240303/57fbee76/attachment.html>