From: Francesco Pietra (chiendarret_at_gmail.com)
Date: Tue Jan 03 2017 - 01:40:20 CST
I went through Example1 with parallel ORCA 3.0.3. As expected from the
standard implementation, increase of core number on a single node resulted
in very little speed up. I have not tried on multi-node but QM codes do not
give much hope. Considering that the QM 20 atoms are of 1st row elements,
at present stage of the QM-MM technique only extremely fast processes could
be investigated, probably no reaction, especially if transition elements
are involved. Or turn to semiempirical.
My question is, did I miss flags to accelerate ORCA's work, such as the
flooding potential of Gromacs or something else? If not, is any plan to
implement some acceleration of the QM part?
This archive was generated by hypermail 2.1.6 : Sun Dec 31 2017 - 23:20:56 CST