[Pw_forum] parallel version for Windows

[Axel Kohlmeyer]

On Sun, Dec 7, 2014 at 11:08 AM, David Foster <davidfoster751 at yahoo.com> wrote:
> Dear Axel
>
> Thanks for your effort for providing the windows version. Until now, I ran the linux version which had been compiled with ifort, intelmpi, icc on a cluster of
>
> CentOs6.5. I downloaded windows version last day, and did according your user guide (installing mpich2, ...).
>
>  I ran sample input for a SMP system with 16 core. It was interesting and very simple. However, I have a question:
>
> Is it possible to use HPC tools to run the program through a windows cluster. What should I do?

i wouldn't know. i don't use windows (except for an occasional quick
test whether the installer packages still do what they are supposed to
do or to debug a reported issue) and don't compile on windows. the QE
windows packages are built on a linux machine using a cross compiler.

you most certainly would have to compile a version yourself that uses
the same MPI library than what is used on the cluster. that MPI
library is most likely compiled against the microsoft compilers, so
you'd have to compile QE natively on that machine, which is likely a
significant porting and debugging effort. the MS compilers are known
to be quite idiosyncratic. in short, it is likely easier to convince
folks to put linux on that cluster.

axel.


>
> Regards
>
> David Foster
>
> Ph.D. Student of Chemistry
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://pwscf.org/mailman/listinfo/pw_forum



-- 
Dr. Axel Kohlmeyer  akohlmey at gmail.com  http://goo.gl/1wk0
College of Science & Technology, Temple University, Philadelphia PA, USA
International Centre for Theoretical Physics, Trieste. Italy.