[Pw_forum] restart from an interrupted phonon run

Paolo Giannozzi giannozz at nest.sns.it
Wed Aug 3 19:19:30 CEST 2005


On Wednesday 03 August 2005 17:23, Stefano Baroni wrote:

> > Note however that
> > 1) restarting is a mess, because files are often corrupted
> > 2) in parallel execution, restarting is a mess multiplied by 1024:
> >   you need to run on the same number of processors, with the same
> >   number of 'pools', and each processor must see all files produced
> >   by all other processors
>
> ma questo problema non era stato rimosso dal restart centralizzato ???

if I remember correctly the phonon code doesn't (yet) store the 
intermediate results needed to restart into a single file

For those of you not especially fluent in Italian, the question
(which was intended to me and not to the forum, of course) was
whether the 'centralisation' of restart files into a single file hadn't
already solved this problem of restarting in parallel

Paolo

-- 
Paolo Giannozzi             e-mail:  giannozz at nest.sns.it
Scuola Normale Superiore    Phone:   +39/050-509876, Fax:-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy



More information about the users mailing list