Thanks W. and Blais.
I ran all six 2-core combinations and I got the same results. I passed
--cpu-set 0,1,2,3 to np2 and I still got the same thing.
I monitored the CPU usage and nothing is out of the ordinary.
The good news is that I don't get this issue when I run on the desktop
which has
In addition to what Wolfgang wrote,
Sometimes the OS will map the cores of a CPU in the following order:
Physical-Logical-Physical-Logical-Physical-Logical etc.
Consequently, if your processor supports hyperthreading through the use of
logical core, running with core 0-1 means you are essentially
On 11/15/23 04:09, Abbas Ballout wrote:
If I run mpirun -np2 the assemble times are 25.8 seconds and the MUMPS
solve times are 51.7 seconds.
If I run mpirun --cpu-set 0-1 -np 2, the assemble times are 26 seconds
(unchanged) but the solve time are at 94.9 seconds!
Is this normal and expecte
This isn't a dealii problem per-se.
I am trying to run a number of simulations with different parameters of the
same code with the underlying solver being MUMPS. I am using *mpirun
--cpu-set * to bind and isolate the different simulations to different
cores (As I believe I should)
*Profilin
