[QE-users] Using h_psi

[Paolo Giannozzi]

Actually h_psi can be applied to whatever set of vectors. In the Davidson
diagonalization, for instance, it is applied to correction vectors having
the form D^{-1}(H-\epsilon)\psi, where D is an easy to invert approximation
of H

Paolo

On Mon, Mar 1, 2021 at 4:25 PM Andrew Xu <andrewaccount at gmail.com> wrote:

> Hi users,
>
> I would like to use the routine h_psi on a bunch of vectors {psi} that are
> not related to each other -- in the code, the psi going into h_psi is
> typically applied to matrices where each column approximates a different
> band, and the subroutines that h_psi calls has variables describing the
> number of columns in psi as related to the number of bands (i.e.,
> max_ibands in vexx_k, for example).
>
> I would like to apply h_psi to a bunch of vectors that are not related to
> each other (i.e., not different bands), and was wondering if anything
> breaks if I make a single h_psi call on a matrix formed by gathering all
> those vectors together.
>
> I would like to avoid calling h_psi on each vector individually because I
> think that is not great for parallelization.
>
> Best regards,
> Andrew
> _______________________________________________
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
> users mailing list users at lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users



-- 
Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
Univ. Udine, via delle Scienze 206, 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/20210301/40bf4a4f/attachment.html>