[gmx-users] domain decomposition error

[Justin A. Lemkul]

nikhil damle wrote:
> Hi,
> 
> It is a vacuum simulation. That's why a molecule can be considered in 
> isolation and i do not need to use periodic box. Secondly, even when i 
> try to run the process on 2 nodes it gives same error. So number of 
> processors might not be the actual reason.
> 

I still disagree.  You have a simulation cell of some size, defined by the box 
vectors in your coordinate file.  Domain decomposition seeks to split this 
volume across whatever number of processors you have defined.  If it cannot meet 
the minimum requirements for a DD cell, it produces the error you are reporting. 
  Check the list archives, this has come up before with DD in vacuo.  You will 
probably have to use mdrun -pd or only use a single processor.  Otherwise, take 
the advice mdrun has printed and adjust the DD parameters to see if you might 
get it to work, although it may simply not be possible.

-Justin

> Regards,
> Nikhil
> 
> ------------------------------------------------------------------------
> *From:* Justin A. Lemkul <jalemkul at vt.edu>
> *To:* Discussion list for GROMACS users <gmx-users at gromacs.org>
> *Sent:* Sun, 11 October, 2009 4:20:58 PM
> *Subject:* Re: [gmx-users] domain decomposition error
> 
> 
> 
> nikhil damle wrote:
>  > Hi,
>  >
>  > I am trying to run a vacuum simulation of a system associated with 
> ATP and peptide. Being vacuum MD i removed pressure coupling and 
> included plain electrostatic cut-offs while calculating potential. 
> Energy minimisation converged to machine precesion even though not upto 
> the specified cut-off limits.
>  >
>  > Now when i am running production MD on multiple processors it gives 
> following error.
>  >
>  > There were 96 inconsistent shifts. Check your topology
>  > There were 96 inconsistent shifts. Check your topology
>  > There were 4 inconsistent shifts. Check your topology
>  > There were 4 inconsistent shifts. Check your topology
>  > There were 6 inconsistent shifts. Check your topology
>  > There were 6 inconsistent shifts. Check your topology
>  >
> 
> These are related to periodic boundary conditions.  Search the list archive.
> 
>  > -------------------------------------------------------
>  > Program mdrun, VERSION 4.0.5
>  > Source code file: domdec.c, line: 5873
>  >
>  > Fatal error:
>  > There is no domain decomposition for 8 nodes that is compatible with 
> the given box and a minimum cell size of 1.025 nm
>  > Change the number of nodes or mdrun option -rcon or -dds or your 
> LINCS settings
>  > Look in the log file for details on the domain decomposition
>  >
>  > Why in the very first place, it is considering box and cell size for 
> a vacuum MD when no related parameters are input ? This error is not given
> 
> If you are running MD, then you most certainly have a box.  How could 
> you have periodicity if your unit cell was not defined?
> 
> The simple fact is that DD cannot split such a small system over 8 
> processors. Either try fewer processors, or switch to particle 
> decomposition with mdrun -pd.
> 
> -Justin
> 
>  > if i run on single processor. What could be the possible reason ? I 
> have checked my topology and it is proper. How should i proceed. Fast 
> reply will be appreciated.
>  >
>  > Regards,
>  > Nikhil
>  >
>  > ------------------------------------------------------------------------
>  > Keep up with people you care about with Yahoo! India Mail. Learn how 
> <http://in.rd.yahoo.com/tagline_galaxy_1/*http://in.overview.mail.yahoo.com/connectmore>.
>  >
>  >
>  > ------------------------------------------------------------------------
>  >
>  > _______________________________________________
>  > gmx-users mailing list    gmx-users at gromacs.org 
> <mailto:gmx-users at gromacs.org>
>  > http://lists.gromacs.org/mailman/listinfo/gmx-users
>  > Please search the archive at http://www.gromacs.org/search before 
> posting!
>  > Please don't post (un)subscribe requests to the list. Use the www 
> interface or send it to gmx-users-request at gromacs.org 
> <mailto:gmx-users-request at gromacs.org>.
>  > Can't post? Read http://www.gromacs.org/mailing_lists/users.php
> 
> -- ========================================
> 
> Justin A. Lemkul
> Ph.D. Candidate
> ICTAS Doctoral Scholar
> Department of Biochemistry
> Virginia Tech
> Blacksburg, VA
> jalemkul[at]vt.edu | (540) 231-9080
> http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin
> 
> ========================================
> _______________________________________________
> gmx-users mailing list    gmx-users at gromacs.org 
> <mailto:gmx-users at gromacs.org>
> http://lists.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at http://www.gromacs.org/search before posting!
> Please don't post (un)subscribe requests to the list. Use the www 
> interface or send it to gmx-users-request at gromacs.org 
> <mailto:gmx-users-request at gromacs.org>.
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
> 
> ------------------------------------------------------------------------
> Keep up with people you care about with Yahoo! India Mail. Learn how 
> <http://in.rd.yahoo.com/tagline_galaxy_1/*http://in.overview.mail.yahoo.com/connectmore>.

-- 
========================================

Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================