On 08/16/2018 04:08 PM, [email protected] wrote:
I'm trying to use periodic boundary conditions with mpi, and the code works
well when I use up to 8 CPUs. However, when I try using more CPUs, the code
crashes with the error:
mpirun noticed that process rank 0 with PID 9529 on node exited on signal 11.
It crashes at the line sparsity_tools::distribute_sparsity_pattern. I went
trough sparsity_tools.cc and it looks like two different strategies are being
used for number of partitions being below or above 8 (recursive and kway), and
was wondering maybe that is related to the source of problem.
I'm able to run the same code with more than 8 CPUs if I don't use periodic
boundary conditions.
Has anyone encountered the same problem? Any suggestions?
Hesaneh,
it's hard to say what the problem is. The way to debug this is to either run
the program in a debugger to see where exactly the segmentation fault happens
(I have a youtube video that explains how to debug parallel programs), or to
come up with a small program that doesn't do anything useful other than
demonstrate the problem. The latter would be independently useful -- do you
think you can remove everything from your program that is not necessary to
demonstrate the problem and send that to us?
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.
