[Wien] lapw2 mpi parallelization limits

[Laurence Marks]

NFS can be very buggy. If you have an old OS, there is a definite
possibility that you may have a broken NFS. This should be gone in
most systems nowadays, and if you check back through the mailing list
there is some discussion of this and was to check. There is a delay
parameter set in parallel_options and if you increase this it might
help or at least give some ideas. (I know of one computer cluster at a
US national lab which I stopped using 12 months ago because it's NFS
was completely broken.)

Probably not relevant thoughts. Do you have root access, so you can
use something more advanced than openmpi (I like mvapich) or can tweak
your nfs setting? Also, was your mpif90 compiled with the same version
of the compiler as you are currently using? I saw somewhere that this
could be a problem.

On Tue, Mar 17, 2009 at 10:07 AM, Scott Beardsley <scott at cse.ucdavis.edu> wrote:
>> Do you have    TOT   or FOR   in case.in2(c) ??
>
> $ cat wientest.in2
> TOT             (TOT,FOR,QTL,EFG,FERMI)
>      -9.0      92.0 0.50 0.05                EMIN, NE, ESEPERMIN, ESEPER0
> TETRA    0.000          (GAUSS,ROOT,TEMP,TETRA,ALL      eval)
>  0 0  2 0  4 0  4 4  6 0  6 4
>  0 0  2 0  4 0  4 4  6 0  6 4
>  0 0  2 0  4 0  4 4  6 0  6 4
>  0 0  1 0  2 0  2 2  3 0  3 2  4 0  4 2  4 4  5 0  5 2  5 4  6 0  6 2
> 6 4  6 6
>  14.00          GMAX
> NOFILE        FILE/NOFILE  write recprlist
> $
>
>> Does it work with a machines file    compute-0-13 compute-0-19 ...; i.e.
>> when you are using different nodes (but restricting the total number to 4)
>
> Yes.
>
>> Could it be that your system does not allow to open a file for reading more than 4 times ??
>
> I can't imagine how. Is WIEN incompatible with NFS? My working directory
> is mounted over NFS.
>
> Scott
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>



-- 
Laurence Marks
Department of Materials Science and Engineering
MSE Rm 2036 Cook Hall
2220 N Campus Drive
Northwestern University
Evanston, IL 60208, USA
Tel: (847) 491-3996 Fax: (847) 491-7820
email: L-marks at northwestern dot edu
Web: www.numis.northwestern.edu
Chair, Commission on Electron Crystallography of IUCR
www.numis.northwestern.edu/
Electron crystallography is the branch of science that uses electron
scattering to study the structure of matter.