From: Giacomo Fiorin (giacomo.fiorin_at_temple.edu)
Date: Sun Sep 26 2010 - 08:38:51 CDT
Hello Cheng, is the simulation going with useGrids set to on or off? Unless
the hills are projected onto a grid, they can quickly become too many to be
computed efficiently.
Also, even if useGrids is on, does the system ever leave the boundaries? In
that case, the program reverts to computing all of the hill's forces off the
grid, which is very slow again. I would say that making the boundaries far
enough from each other so that the variable is always inside the grid solves
the problem.
Bests
Giacomo
---- ----
Dr. Giacomo Fiorin
ICMS - Institute for Computational Molecular Science - Temple University
1900 N 12 th Street, Philadelphia, PA 19122
giacomo.fiorin_at_temple.edu
---- ----
2010/9/26 OiniL <noinil_at_gmail.com>
> Dear Namd Users:
>
>
> I'm using metadynamics to simulate the folding of a protein with NAMD.
>
> After several hours... Running speed got more and more slower.
>
> The speed at the beginning was
> "CPU: 2.30465, 0.00455531/step Wall: 2.30378, 0.00455352/step,"
> but 10 hours later, it's
> "CPU: 165525, 0.0223486/step Wall: 165964, 0.0223666/step,"
>
> about 1/5 the speed at the beginning.
>
> I tested another equilibrium run with the conf.file totally same as the
> metadynamics file except the part of colvars, and this time the speed looks
> fast and stable.
>
> Is there anybody else have had a same experience?
> How can I check what is wrong?
>
> Any help or hint is appreciated
> Best regards!
>
> Cheng TAN
> PhD student of Biophysics Dept.
>
> Nanjing University
>
> --
>
> ---------------------------------------------------------------------------------
> ×£
> ½¡¿µ¿ìÀÖ
>
> Ì·Ø©
>
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:54:33 CST