Re: expected speed up after fixing atoms?

From: Kenno Vanommeslaeghe (kvanomme_at_rx.umaryland.edu)
Date: Wed Sep 24 2014 - 14:06:46 CDT

The best person to answer that question is you. Speed of a simulation can
never be known without actually running it, and you'll waste more time by
trying to pry an answer out of the community than by running a couple of
quick test calculations.

That said, I have to agree with Aaron I wouldn't expect the speedup to be
spectacular, if at all. This enormously depends on implementation details,
but even if NAMD would be implemented to maximally profit from fixed atoms
(which it isn't), fixing half of the atoms in the system can never get you
more than ~25% speedup, which would roughly correspond to a ~15%
improvement in conformational sampling for the CPU time - typically not
worth introducing coarse approximations into the system...

On 09/24/2014 01:44 PM, Jose Borreguero wrote:
> Dear NAMD users,
>
> I have a system where I can fix the positions of more than half the atoms
> without affecting much the dynamics I'm interested in. I was wondering
> what is (approximately) the expected speed up in the simulation after I do
> this. Even more important: What will happen if I run a parallel job with
> this system? Will some of the CPU end up doing no calculations if the
> fixed atoms are assigned to those CPU's?
>
> Best regards,
> Jose Borreguero

This archive was generated by hypermail 2.1.6 : Thu Dec 31 2015 - 23:21:15 CST