[Q-e-developers] Quantum Espresso: data distribution and work distribution

[Paolo Giannozzi]

Your questions are too generic. QE implements several levels of
parallelization, more or less described in J.Phys.:Condens.Matter
21, 395502 (2009), and in other places like here:
http://www.fisica.uniud.it/%7Egiannozz/QE-Tutorial/tutorial_para.pdf
The data distribution in QE in particular is a rather complex story.
Most arrays are distributed, some in simple ways, some in not-so-simple
ways

Paolo

On Mon, 2015-03-09 at 16:27 +0100, Federico Padua wrote:
> Dear Quantum Espresso team,
> 
> I am not a user of QE but we are running it to evaluate some
> checkpointing related behavior. I am interested in the following
> information about QE since I'm writing part of the publication (a paper)
> and I would like to get the right info directly from the developers. We
> are using version > 4.2 . I browsed the developers' docs but still some
> simple points are not clear. I have the following questions:
> 
> - how Quantum Espresso distribute the data to be processed among processors?
> - how Quantum Espresso distribute the work (task if you will) among the
> processors involved in the calculation?
> 
> 
> Thank you very much in advance for your answer and for your time.
> 
> 
> Best Regards,
> 
> Federico
> _______________________________________________
> Q-e-developers mailing list
> Q-e-developers at qe-forge.org
> http://qe-forge.org/mailman/listinfo/q-e-developers

-- 
Paolo Giannozzi, Dept. Chemistry&Physics&Environment, 
Univ. Udine, via delle Scienze 208, 33100 Udine, Italy
Phone +39-0432-558216, fax +39-0432-558222