('binary' encoding is not supported, stored as-is)
Dear Dr. Walker,
I ran MD for 100 steps for 2 of the minimized structures on 1,2 and 4
processors on the cluster and for 1 and 2 processors on Octane.
The total energy (ETOT) was exactly the same in the output for all 5
processor combinations.
For the first structure it was 545.8669 (kcal/mole), not sure about the
units, but that was the value it showed in the ETOT place of the output file.
For the second structure it was 568.2084, for all outputs.
As over the first 100 steps the values were exactly the same, when coming
from the same input structure, the installations seem to be ok and the
problem (although it isn't really one, as the MD seems to work) is just
that my structure has a hard time converging on a minimum energy during
minimization. Is this correct, or did I misunderstand something?
Thank you for helping me through understanding this!
Best Regards,
Armin
==============Original message text===============
On Wed, 25 Aug 2004 18:35:18 EDT "Ross Walker" wrote:
Dear Armin,
> I checked the first two of your suggestions already.
> No, shake was not turned on.
> Secondly, I went back and ran the tests again. They were fine with 2
> processors, only one that said possible failure. The same was
> true for 1
> processor. The output there was as follows:
> To run it with 1 processor I have to use mpirun -np 1, would
> this make a
> difference somehow?
Only in that you are stil running the MPI version of the code. Just not
actually doing any broadcasts. If the tests all pass ok though then you are
probably ok. It is possible that something is wrong with your system such
that it cannot be minimised well. E.g. if you have two atoms very close
together they will generate a very large initial force. The minimiser will
then take a big step to try and get away from this. It is possible that if
this force is very large you will get radically different behaviour from the
minimiser on different architectures. I would simply try doing some MD with
your minimised structure. If it rapidly blows up (watch it in VMD) then you
most likely have a problem with your molecule.
> I will try your other suggestions soon, I just wanted to know
> the answer
> to these questions sooner.
> As I do not know if the Amber installation on the cluster has
> passed the
> tests, would it make more sense to contact that administrator and ask
> about the tests before performing the short MD runs you suggested?
You could do. It won't take long to run the quick MD runs though and it will
let you know if there is a problem with your molecule. I suspect that what
you have been seeing is an anomally unique to your current initial molecular
configuration and has been fooling us into thinking there is something wrong
with your amber installation. Try out the DNA tutorial simulations and see
if you get similar results for the short MD runs.
Let me know what happens to the MD simulation of your molecule. If the sims
from the DNA tutorial are ok but your molecule blows up then it will confirm
it is a problem with your molecule and not your amber installation. If both
are fine then I give up... ;-)
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
Received on Thu Sep 02 2004 - 01:15:11 PDT
