[Pw_forum] repeated band structure calculation in splittet phonon calculation

Paolo Giannozzi giannozz at democritos.it
Tue Aug 2 18:54:26 CEST 2011


On Aug 1, 2011, at 13:32 , Thomas Gruber wrote:

> In a phonon calculation with a different k-point grid than in a scf
> calculation a band structure calculation is needed for every q-point,
> but not for every irreducible representations. Now I want to split  
> this
> phonon calculation in every q-point and every irreducible
> representations. But when I do that, there will be a band structure
> calculation in every job, but only one per q-point is necessary.
> How can I save the band structure calculation to be read for every  
> job?

In principle you could just copy the results for the needed band  
structure,
but I am not sure the code will recognize that the data is there. You  
may
need to modify the code. Keeping track of all possible cases is not  
easy.
Typically the calculation of bands is a small part wrt a phonon  
calculation.
The ultimate solution is to speed up the calculation of bands, but  
this would
require some ideas and some serious work

> Or did I something wrong and it should work?

what should work in which case and with which version of the code  
exactly?

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







More information about the users mailing list