Dear Thomas,
1. Do you compute on the same mesh?
>
Yes, it is exactly the same mesh which is illustrated within your paper:
16384 cells (50115 DoFs). You can observe the same value also in the
terminal output. I selected 4 global pre-refinement steps and 2 adaptive
refinement cycles.
> 2. do you compile in release mode or debug mode?
> Release is faster.
>
Release mode.
3. Possibly you can optimize deal.Il with respect to the CPU etc.
On my desktop computer at work, deal.II is 2 - 4 times slower than
on my laptop for instance.
I am using an Intel® Core™ i7-4770 CPU @ 3.40GHz × 8 processor. Hence, I
assumed it would be even faster than your Intel(R) Core(TM) i5-3320M
CPU@2.60 GHz machine. What MPI solver did you use? Intel with OpenMPI?
Since I am using MPICH maybe there are some differences. I mean the
difference in speed is not magic and should be based on the processor type
or the settings of the solver. Maybe I should check how fast my MPI solver
works outside of deal.II in order to see where the problem lies.
Best regards,
Seyed Ali
--
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.
