no, standard remd requires all to run at the same time. otherwise how will
the new set of replicas exchange during MD with the set previously run?
there is a way to do such a thing but it's an advanced method (reservoir
REMD) and should not really be used until you are quite familiar with how
REMD works.
even though you only have 16 processors you can still run all of your
replicas at one time.
On Thu, Sep 27, 2012 at 1:46 AM, gargi borgohai <gargib2011.gmail.com>wrote:
> Dear AMBER users,
> I am trying to perform REMD simulation of a peptide and water
> over the temperature range 274-480K distributed among 40 replicas. I want
> to ask if I have only 16 processors then can I perform REMD simulation for
> the 40 replicas not simulating all of the replicas at a time but in three
> steps i.e. if i simulate first 16 replicas followed by next 16 replicas and
> finally the last 8 replicas. Will it cause any difference while analysing
> the results ( regarding the exchange criteria ) in comparision to those
> obtained from REMD simulation of all the replicas at a time ( i.e.when 40
> replica is simulated at 40 processors )?
>
> Thanking you.
> Sincerely
> Gargi Borgohain.
> IIT Guwahati.
> _______________________________________________
> 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
Received on Thu Sep 27 2012 - 03:30:03 PDT
