[gmx-users] problem regarding cut-off and the PME grid spacing

Mark Abraham mark.j.abraham at gmail.com
Wed Apr 20 19:09:23 CEST 2016


Hi,

On Wed, 20 Apr 2016 19:04 Tuhin Samanta <samantatuhin1991 at gmail.com> wrote:

> Thank Mark for your suggestion.
>
> I was trying to resolve the problem --------
> " 4 particles communicated to PME node 4 are more than 2/3 times the
> cut-off out of the domain decomposition cell of their charge group in
> dimension x.
> This usually means that your system is not well equilibrated."
>
> First of all I have done the energy minimization for long times for proper
> equilibriuation.
>

Then what did you change? You may need to be more gentle.

Secondly by changing rcut and PME mesh girding I have tried a lot. But it's
> showing same error message.
>

Those suggestions are made for perhaps improving performance, which is
unrelated to your error. They just both happen to use the word PME...

So can you please tell me any specific solution for this problem.
>

They're all at http://www.gromacs.org/Documentation/Terminology/Blowing_Up

Mark

Thanking you,
> Tuhin
>
>
> On Wed, Apr 20, 2016 at 2:21 PM, Mark Abraham <mark.j.abraham at gmail.com>
> wrote:
>
> > Hi,
> >
> > The information message and the actual error are separated in your log
> file
> > with a bunch of punctuation because they're logically separate.
> >
> > Anyway, there should also have been a link to
> > http://www.gromacs.org/Documentation/Errors which suggests some
> strategies
> > for this issue.
> >
> > Mark
> >
> > On Wed, 20 Apr 2016 09:00 Tuhin Samanta <samantatuhin1991 at gmail.com>
> > wrote:
> >
> > > Hello ,
> > >
> > > My system involves water molecules in a box of 5x5x5 nm3. When I was
> > doing
> > > the simulation for energy minimization I  got n error message --
> > >
> > > The optimal PME mesh load for parallel simulations is below 0.5
> > >   and for highly parallel simulations between 0.25 and 0.33,
> > >   for higher performance, increase the cut-off and the PME grid spacing
> > >
> > >  4 particles communicated to PME node 4 are more than 2/3 times the
> > cut-off
> > > out of the domain decomposition cell of their charge group in dimension
> > x.
> > > This usually means that your system is not well equilibrated.
> > >
> > > I have changed the cut off and grid spacing but still it's showing the
> > > error message. I know it is the problem with the parallelization. Also
> I
> > > have  gone through the user forum's discussion.  So can anyone please
> > > suggest me how to get rid of that problem.
> > >
> > >
> > > Thank you so much.
> > > Best regards,
> > > Tuhin
> > > --
> > > Gromacs Users mailing list
> > >
> > > * Please search the archive at
> > > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> > > posting!
> > >
> > > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> > >
> > > * For (un)subscribe requests visit
> > > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> > > send a mail to gmx-users-request at gromacs.org.
> > >
> > --
> > Gromacs Users mailing list
> >
> > * Please search the archive at
> > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> > posting!
> >
> > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> >
> > * For (un)subscribe requests visit
> > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> > send a mail to gmx-users-request at gromacs.org.
> >
> --
> Gromacs Users mailing list
>
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> posting!
>
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> send a mail to gmx-users-request at gromacs.org.
>


More information about the gromacs.org_gmx-users mailing list