[Pw_forum] Problems with 'relax' calculations

Giovani Faccin faccin.giovani at gmail.com
Wed Feb 16 14:28:25 CET 2011


Ettore,

One thing that can make pw.x freeze is bad behaviour from the MPI
implementation<http://www.quantum-espresso.org/user_guide/node58.html#SECTION000125030000000000000>.
I've experienced random freezes myself when using QE with openmpi on a
single node, multicore computer.

If you are running your calculation in parallel, give it a try in serial
just to see if it can pass the point where this error is occuring.

Giovani Faccin

CPPP-UFMS/IFGW-Unicamp - Brazil



2011/2/16 Ettore Baldini Neto <neto.baldini at gmail.com>

> Dear users,
>
> I'm trying to relax a 32C graphene layer with a Ti atom and the calculation
> simply stops without any error messages.
> I'm attaching my input  in order to get some insight in what's going on. Is
> there any parameter badly defined?
> Although I know I must include spin polarization, for now I'm not
> considering it. But  what would it be a good value
> for starting_magnetization for Ti?? I have no experience in performing such
> calculations.
>
> Thanks for your help
>
> Bests
>
> Ettore
>
> PS.  The last thing it writes in the output is:
>
> .....
> iteration #  4     ecut=    40.00 Ry     beta=0.70
>     CG style diagonalization
>     ethr =  1.95E-07,  avg # of iterations =  3.2
>
>     negative rho (up, down):  0.369E-02 0.000E+00
>
>     total cpu time spent up to now is   8905.97 secs
>
>     total energy              =    -481.99306649 Ry
>     Harris-Foulkes estimate   =    -481.99307188 Ry
>     estimated scf accuracy    <       0.00002695 Ry
>
>     iteration #  5     ecut=    40.00 Ry     beta=0.70
>     CG style diagonalization
> ..........
>
>
> **********************************************************************
> Dr. Ettore Baldini-Neto
> Researcher
> W.von Braun Center for Advanced Research, Campinas, Brazil
> **********************************************************************
>
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
>


-- 
Giovani
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.democritos.it/pipermail/pw_forum/attachments/20110216/d01ff9e7/attachment-0001.htm 


More information about the Pw_forum mailing list