[gmx-users] what's the maximum number of atoms that GROMACS can simulate?

[Kutzner, Carsten]

Hi Zhang,

> On 9. Jul 2019, at 15:16, 张驭洲 <zhangyuzhou15 at mails.ucas.edu.cn> wrote:
> 
> Hello，
> 
> I want to know if there is a maximum number of atoms that GROMACS can simulate, or if there are any bugs in the code that cause the error in computing the coordinate of atoms when the number of atoms is very large? I'm trying to simulate some very large systems using GROMACS-2019.3. Unfortunately, the energy minimization crashed during the run. I tried using more than one compute nodes, for example, 4, 8 or 32, and I'm sure the memory is not exhausted. The system I tried contains more than 100 million atoms.
There is no limit on the number of atoms that you can simulate in GROMACS.

However, if the simulation systems become very large in at least one dimension,
you might need double precision, as single precision could be insufficient 
to accurately store the positions of atoms that are far off the origin.
This can then lead to a positive energy drift which could become so large
that the system gets instable.

A while time ago we have observed such problems with large systems of mostly
vacuum, which contained several water droplets. The droplets that were farthest
away from the origin would start to heat up and eventually evaporate after 
some time, unless a double precision version of mdrun was used.

Carsten


> 
> Zhang
> -- 
> 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.

--
Dr. Carsten Kutzner
Max Planck Institute for Biophysical Chemistry
Theoretical and Computational Biophysics
Am Fassberg 11, 37077 Goettingen, Germany
Tel. +49-551-2012313, Fax: +49-551-2012302
http://www.mpibpc.mpg.de/grubmueller/kutzner
http://www.mpibpc.mpg.de/grubmueller/sppexa