[Pw_forum] ph.x: Avoiding the recalculation of the band structure in distributed phonon dispersion jobs

[Karttunen Antti]

Dear Andrea,

Thank you, this solution works perfectly for us! Now the whole process goes really smoothly.

Best wishes,
Antti

-- 
Dr. Antti Karttunen
Department of Chemistry
University of Jyväskylä, Finland
Tel: +358-50-3473475
WWW: http://www.iki.fi/ankarttu 

-----Original Message-----
From: pw_forum-bounces at pwscf.org [mailto:pw_forum-bounces at pwscf.org] On Behalf Of Andrea Dal Corso
Sent: Wednesday, February 27, 2013 3:30 PM
To: PWSCF Forum
Subject: Re: [Pw_forum] ph.x: Avoiding the recalculation of the band structure in distributed phonon dispersion jobs

I am not going to implement this in the SVN version, at least not now.
However it seems that if you reopen the wavefunctions after saving them
with twfcollect=.true.. with something like:

     CALL punch( 'all' )
     IF (only_init) THEN
        CALL clean_pw( .TRUE. )
        CALL close_files(.true.)
        wfc_dir=tmp_dir_phq
        tmp_dir=tmp_dir_phq
        CALL read_file()
        IF (.NOT.lgamma_iq(iq).OR.(qplot.AND.iq>1)) CALL
set_small_group_of_q(nsymq,invsymq,minus_q)
     ENDIF

you can both run the epsilon calculation and the next ph.x runs with a
different number of processors. It is really inelegant, and I think
there are better ways to do this, but it seems to work.

Best wishes,

Andrea