Yeah, that method of writing the restarts is what I had started to use
recently. It might still be nice if the default behavior was to have an
old and a new restart file, that way you wouldn't build up a big heap of
them over the course of a multiple day run for instance. But they are
fairly small, so I guess it's really a minor issue and doesn't exactly need
a change.
On Tue, Feb 14, 2012 at 10:15 PM, filip fratev <filipfratev.yahoo.com>wrote:
> You are right. This
> is only for "emergency" situation, thus no problem.
> Thank you again!
>
> All the best,
> Filip
>
>
> ________________________________
> From: case <case.biomaps.rutgers.edu>
> To: filip fratev <filipfratev.yahoo.com>; AMBER Mailing List <
> amber.ambermd.org>
> Sent: Wednesday, February 15, 2012 4:52 AM
> Subject: Re: [AMBER] Restart file for pmemd not showing all information
>
> On Tue, Feb 14, 2012, filip fratev wrote:
>
> > It works for me and if I set in my input file ntwr to -10
> > 000 than I got "full" restart files on each 10 000 steps. However, it
> > continuously saves restart files on each 10 000 steps, i.e.
> MyRes.rst_10000,
> > MyRes.rst_20000 and so on. The only problem is that I will obtain 1000
> restart
> > files for a 100ns simulation if I set ntwr=-50 000.
>
> How often do you need such restarts? I thought this was only for an
> "emergency" (rare?) situation where the job crashes before finishing. If
> you
> save the restart file every 12 hours or so, you won't loose more than 12
> hours
> of computation, etc. I guess it's a tradeoff between (temporary) disk
> space,
> and potential loss of time.
>
> ....dac
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
Aron Broom M.Sc
PhD Student
Department of Chemistry
University of Waterloo
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Feb 14 2012 - 20:00:02 PST
