[Pw_forum] PWscf and highly parallel machines

[Nichols A. Romero]

If one were to use CP across such a large number of processors, what
algorithm is your best bet
for minimizing your electronic d.o.f., sd or cg


On 2/14/07, Paolo Giannozzi <giannozz at nest.sns.it> wrote:
>
> On Feb 14, 2007, at 10:39 , Alain Allouche wrote:
>
> > I wonder if somebody have the experience of PW codes
> > running on highly parallel machines like in those of the DEISA
> > supercomputing grid, are the codes like pw.x supporting
> > such an environment ?
>
> they have to, since I am involved in a DEISA project that
> will use pw.x :-) .Support for very large calculations with
> pw.x is less mature than for the Car-Parrinello code, which
> on test run has reached 4000 processors or so. With PW
> codes some work will be needed in order to perform really
> large calculations (notably, reduction of nonscalable memory
> usage and of nonparallel parts of the code) but it will be
> done.
>
> Paolo
> ---
> Paolo Giannozzi, Democritos and University of Udine, Italy
>
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>



-- 
Nichols A. Romero, Ph.D.
1613 Denise Dr. Apt. D
Forest Hill, MD 21050
443-567-8328 (C)
410-306-0709 (O)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20070214/23ae49ff/attachment.html>