[Pw_forum] Serial vs. parallel runs of berry.in test case

[Paolo Giannozzi]

Actually it seems related to nonuniform division of the FFT grid across
processors: with 8 processors it works for fft grid=32x32x32, not for
36x36x36. Definitely a bug

Paolo

On Tue, Aug 18, 2015 at 9:15 AM, Paolo Giannozzi <p.giannozzi at gmail.com>
wrote:

> It is related to the real-space treatment of augmentation charge (option
> tqr=.true.). It works with 2 and 4 processors but not on 8. It is a rather
> extreme case (8 processors for a rather small system) but this shouldn't
> happen.
>
> Paolo
>
> On Tue, Aug 18, 2015 at 7:29 AM, Mostafa Youssef <myoussef at mit.edu> wrote:
>
>> Dear all,
>>
>> I tried to run the standard tests of PWscf  both in series and in
>> parallel. In series all is fine.  In parallel, I found the case of
>> berry.in to be problematic as I got an error:
>>
>>  Error in routine addusdens_r (1):
>>  WRONG CHARGE expected  44.000000, found  44.404701: ions may be
>> overlapping or increase ecutrho
>>
>> And the code stops at the first scf step. When I double the energy and
>> charge density cutoffs , it works  well in parallel. This behavior is
>> reproduced in versions 5.1.1, 5.1.2, and 5.2.0.
>>
>> Any thoughts on the possible reasons are appreciated.
>>
>>
>> Thank you,
>> Mostafa Youssef
>> MIT
>>
>> _______________________________________________
>> Pw_forum mailing list
>> Pw_forum at pwscf.org
>> http://pwscf.org/mailman/listinfo/pw_forum
>>
>
>
>
> --
> Paolo Giannozzi, Dept. Chemistry&Physics&Environment,
> Univ. Udine, via delle Scienze 208, 33100 Udine, Italy
> Phone +39-0432-558216, fax +39-0432-558222
>



-- 
Paolo Giannozzi, Dept. Chemistry&Physics&Environment,
Univ. Udine, via delle Scienze 208, 33100 Udine, Italy
Phone +39-0432-558216, fax +39-0432-558222
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20150818/68b9424f/attachment.html>