On Mon, 2014-11-03 at 10:52 -0500, IBRAHIM M ABDELRADY wrote:
> Hi,
>
>
> I am trying to do an MD simulation of
> a protein model using AMBER 12 using the following input for the equilibration
> step:
> The job is running using pmemd on a local cluster using 64 cpu. The run seemed to be OK; however,
>
>
> looking at the output file I see a strange behavior. Some NSTEPs are written more than once and some are missing from the output file.
>
>
> I extracted the "NSTEP" from the output (see below) and you can see that after NSTEP = 10500 ps, the program skipped NSTEP = 11000 ps
>
>
> and recorded NSTEP = 11500 followed by another NSTEP = 10500 ps. The same behavior is shown for other recorded NSTEP.
>
>
> The same behavior happened in the heating step but the final coordinate file seemed to be OK and the time was recorded correctly at the header.
>
>
> Not sure what might be wrong with the run? Any idea what is going on?
>
My guess is you are trying to run pmemd in parallel, but you are using
"pmemd" instead of "pmemd.MPI". Of course since you didn't tell us how
you actually tried to run the command, this is only a guess (but one
that is consistent with what you've seen).
Another possibility is that you are using a different MPI to run the
simulation than the one used to compile Amber.
HTH,
Jason
--
Jason M. Swails
BioMaPS,
Rutgers University
Postdoctoral Researcher
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Nov 03 2014 - 08:00:03 PST
