On 12/10/2017 08:38 PM, Phani Motamarri wrote:
I was just wondering in a MPI run, is there a critical number of
finite-elements to be present in the base mesh on each processor when doing an
adaptive refinement of the base mesh using p4est in deal ii. I had a base
uniform mesh of 64 elements and when I try to do adaptive refinement using
triangulation.execute_coarsening_and_refinement(), the code gets stuck at this
function call when I run more than 8 processors. However, if the base mesh has
512 elements, the code gets stuck at the same function call when I run on 16
processors.
No, there is no critical number. We have computed cases with one coarse mesh
cell and 10,000 processors, and with 30,000 coarse mesh cells and one processor.
You'll have to figure out in more detail what "stuck" concretely means. For
example, are all processors at the same location of your code, or are one or
more waiting in one location and the rest in another (the classical case of a
"deadlock").
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].
For more options, visit https://groups.google.com/d/optout.
