[Pw_forum] band parallelization in QE

Stefano Baroni baroni at sissa.it
Sat May 26 13:10:46 CEST 2012

On May 26, 2012, at 8:49 AM, jiayudai wrote:

> Dear developers and users,
> It's really a good news that there is band parallelization for EXX calculations. Does this function work for the other models in QE? I mean, can we use the same way to accelerate the calculations especially we have large number of bands.

EXX/hybrids require a double loop over bands. (the action of the Fock exchange over a single band requires a number of FFTs equal to the number of occupied bands)

> Besides, is there any difference between EXX and hybrid functional? It is a little puzzled for me because i thought they are difference theories before.

different theories, but essentially identical technical/algorithmic details, as an even superficial study of the theories would indicate.


Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center - Trieste
http://stefano.baroni.me [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)

La morale est une logique de l'action comme la logique est une morale de la pensée - Jean Piaget

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20120526/b075bcbe/attachment.html>

More information about the users mailing list