[QE-users] how to choose the number of processors (-np with -npool)

[Giovanni Cantele]

Dear Bouafia,

you might look for parallelization hints  on the website. The correct use
of the available resources depends on the architecture of your computer and
on the size of the calculation.

You should first understand if the result you find in terms of
computational time are due to the fact that the size of your system is too
small (I would not use 18 processors for bulk silicon, for example)
or to the fact that the two nodes are not efficient in terms of inter-node
communication.

This being said, if 18 processors are a good starting point, you might try
with 36 with -npool 2. This will partition your 36 processors into two
groups of 18 processors, each dealing with one half of the k-points.
This will work only if your calculation comprises at least two k-points.
Giovanni

-- 

Giovanni Cantele, PhD
CNR-SPIN
c/o Dipartimento di Fisica
Universita' di Napoli "Federico II"
Complesso Universitario M. S. Angelo - Ed. 6
Via Cintia, I-80126, Napoli, Italy
e-mail: giovanni.cantele at spin.cnr.it <giovanni.cantele at spin.cnr.it>
Phone: +39 081 676910
Skype contact: giocan74

ResearcherID: http://www.researcherid.com/rid/A-1951-2009
Web page: https://sites.google.com/view/giovanni-cantele/home


Il giorno lun 20 mag 2024 alle ore 13:26 H. BOUAFIA <
hamza_bouafia at hotmail.com> ha scritto:

> Hi
> For two nodes of 18 processors each, how to choose the number of
> processors (-np with -npool) because the calculation with a single node
> (-np 18) is much faster than that with two nodes (-np 36).
> Sincerely
> H Bouafia
> _______________________________________________
> The Quantum ESPRESSO community stands by the Ukrainian
> people and expresses its concerns about the devastating
> effects that the Russian military offensive has on their
> country and on the free and peaceful scientific, cultural,
> and economic cooperation amongst peoples
> _______________________________________________
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
> users mailing list users at lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20240520/98f28e7e/attachment.html>