Re: Time of MD

From: Peter Freddolino (petefred_at_ks.uiuc.edu)
Date: Fri Jun 30 2006 - 11:11:03 CDT

It is obviously possible for this to occur (since you've seen it ;-),
but probably indicates poor network performance between your PCs or
something along those lines. Are both pcs equivalent? And could you post
the namd log files for both cases?
Peter

edu wrote:
> Hello
>
> When I run a MD simulation on 1 PC (using namd) the time involve in
> the process is lower than if I use 2 PC with Charmrun soft. żIt is
> possible?
>
> The simulation is:
>
> A protein in a water box (9250 total atoms). The PCs are connected in
> a window network but the time is :
>
> using 1 PC(localhost): WallClock: 497.609009 CPUTime: 497.608002
> Memory: 0 kB
> using 2 PC(Windows Network): WallClock: 562.856018 CPUTime:
> 562.854980 Memory: 0 kB

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:42:17 CST