On 09/27/2017 09:49 PM, Artur Safin wrote:
The only idea I can come up with is to associate refinement with a
physical point in the domain. After solving the elasticity eq-n, I would
cycle through the coarsest level of a triangulation, and store the
cell->center() point with the cell's refinement history into some file.
For the second simulation on the global mesh, each cell would have to
find its counterpart from the first simulation by making sure their
center points match, and then load in the refinement. While possible,
this seems a bit like an overkill.
So I am curious if there is a better alternative than what I'm thinking
of. Any suggestions?
A better quantity to compare is cell->id() that returns a CellID object
that is globally unique in a parallel distributed triangulation and that
avoids the comparison of floating point numbers you have when looking at
the center of cells.
In general, I think Daniel is right that if you can avoid transferring
data between parallel distributed meshes, then that is the way to go.
It's just too cumbersome and inefficient to do so.
Best
W.
--
------------------------------------------------------------------------
Wolfgang Bangerth email: bange...@colostate.edu
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 dealii+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
