Clearly, omp_global:12 is NOT efficient for lapw1. The diagonalization
does not scale for so many cores.
Otherwise it means that your mpi-installation (what are you using ?
ifort, gfortran, which mpi, blas, scalapack, elpa, ...) is VERY bad.
On 8/27/20 1:44 PM, karima Physique wrote:
Dear Wien2k users,
A calculation with 38 equivalent atoms (96 inequivalent  ), we found
that using lapw1 with omp_global is very faster than lapw1_mpi
machine file 1
omp_global: 12
1: node1
machine file 2
1: node1:12
Have you any comment on this?
_______________________________________________
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
--
P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-165300 FAX: +43-1-58801-165982
Email: bl...@theochem.tuwien.ac.at WIEN2k: http://www.wien2k.at
WWW: http://www.imc.tuwien.ac.at/TC_Blaha
--------------------------------------------------------------------------
_______________________________________________
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
