[Pw_forum] Fwd: Final iteration often crashes after structure optimization

Cohen, Ronald rcohen at carnegiescience.edu
Mon Aug 10 14:01:43 CEST 2015


Lowering ecutrho makes things worse not better. But I understand the
problem better. It is a problem with load balancing. This problem only
arises when R & G space division>1 . With R & G space division=1 it never
crashes in this way. However, the performance with R & G space division=4
is astounding compared with R & G space division=1. I have 16 k-points, yet
with npool=16 it takes 74 seconds for the first k-point, and with nppol=4
on 16 processors (R & G space division=4) it takes 16 seconds--a speedup of
4.6 with the same number of processors! Yet 20% of the time or so R & G
space division>1 fails, presumably because of a load balancing problem. The
solution is to rebalance the R & G space divisions. Is there a developer
out there familiar with this?

Sincerely,

Ron


---
Ronald Cohen
Geophysical Laboratory
Carnegie Institution
5251 Broad Branch Rd., N.W.
Washington, D.C. 20015
rcohen at carnegiescience.edu
office: 202-478-8937
skype: ronaldcohen
https://twitter.com/recohen3
https://www.linkedin.com/profile/view?id=163327727

On Mon, Aug 10, 2015 at 6:15 AM, Youssef <aharbil at gmail.com> wrote:

> Dear Ronald,
>
> Strange behavior, it happens here (ggen.f90)  :
>
>  IF (ngm > ngm_g) CALL errore ('ggen', 'too many g-vectors', ngm)
>
> I am not QE developer, but firstly I suggest to lower ecutrho to 280 Ry
> and see!!!
>
> Youssef Aharbil,
> Laboratory of Physics and Chemistry of Materials
> Faculty of sciences Ben msik, Casablanca
> Morocco
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20150810/5efe9a61/attachment.html>


More information about the users mailing list