[gmx-users] A bug for MPI of GROMACS?

[Mark Abraham]

wuxiao wrote:
> Dear users, 
> 
>   Recently, I encounter a very strange thing while perfoming MD using GROMACS. With the same minimization parameters, mdrun can generate different configuration in parallel way or not. It seems that the former one leads to obvious exploding of the configuration into parts whereas the latter one can obtain a reasonable configuration. Noted that the system is a polymer (dendrimer) without pbc. I wonder if this is a bug for MPI of GROMACS. Could you give me some hints? Thanks a lot for your attention!

This seems wildly unlikely to be the consequence of a bug. Much more
likely is that you haven't done what you think you've done. If you can
write down your steps with complete command lines, and reproduce your
findings, then you're in with a chance of suggesting a bug :-) You'd
need to provide a full description of your workflow, and which GROMACS
version.

Mark