sorry I thought the execution of PW went fine.
It seems you have compiled the program for serial execution and then
linked to parallel libraries.
you should configure for parallel execution and recompile the codes.
if it doesn't work, could you send me the make.inc file for checking ?
On Thu, Oct 22, 2020 at 10:17 AM Thanh-Nam Huynh
wrote:
> When I set OMP_NUM_THREADS=64 and ran
> mpirun -np 2 ...
> in the output, the first lines were printed out twice.
>
Meanwhile, when I set OMP_NUM_THREADS=1 and ran
> mpirun -np 40 ...
> every line was printed out 40 times in the output
Dear Pietro and Marcelo Albuquerque,
I followed your instructions and the job has done greatly. However. there
is something weird in the output file. When I set OMP_NUM_THREADS=64 and ran
mpirun -np 2 ...
in the output, the first lines were printed out twice.
Meanwhile, when I set
Hi, Thanh-Nam Huynh and Pietro.
I agree with what Pietro wrote.
But, please correct me if I'm wrong, I think that in a workstation like
yours you should set OMP_NUM_THREADS=1, and run QE programs (compiled in
parallel) as, for example,
mpirun -np 40 pw.x -inp pw.in > pw.out
Best wishes,
*
Hi
no it shouldn't take that long.
According to the output you should be running on 5120 processors,
that would indeed be a terrific workstation :-).
the openmp version of the code is a little bit tricky, if the number of
available threads is not specified it tries to as many as it
Dear colleagues,
I ran a pp.x job in a workstation using 40 processors to get the potential
of a surface slab a month ago. Despite that all the processors are running
at 100%, the output file stops at these lines, and nothing has changed
during this month.
Writing data to file