On Thursday, January 26, 2012 2:16:21 AM UTC-5, Chetan Mahajan wrote:
Hi
Is it scientifically ok to run cp2k on number of cores more than total number of atoms in the system? If ofcourse problem scales well with that many cores, which is perhaps happening in my system
please keep in mind that for quantum chemistry
calculations, in the crudest approximation, the
basic unit of work is an electron, not an atom.
atoms are for the most part just providing an
"external" field for the electrons to be in.
axel.