[gmx-users] why i cannot the long time md run

[Anton Feenstra]

xiaoyi li wrote:
> To whom it concerns,
> 
> I am trying to run a long time MD simulation on a complicated system
> composed of two axt molecules and two proteins with Gromacs. I set 
> 
> nstxtcout=100
> nstxout=5000
> nstvout=5000
> dt=0.002
> nsteps=500000
> nstlog=200
> 
> in the .mdp file to get the 1000ps result. But the real output result is
> only 110s. The size of the output file are:
> 
> .log  2147483647
> .trr   13120992
> .tpr    6588032
> 
> Is it because of the size limit of the output file? what's the size
> limit of all the output files? what should i do then? If not, why I
> cannot get the right long time run?

I don't know why your output is not complete, but I would suggest writing
less often, especially to the logfile. All energy values are also stored
in the energy file (and much more compact). I set nstlog to 5000 or 10000.
(saves you a factor 20-50!) Also, you are writing 5 frames per ps to your
.xtc trajectory, that is a lot. I usually only write 1 frame per ps (saves
you a factor of 5).

Did you check the logfile and/or screen output of mdrun for any clues why
you simulation has stopped?


-- 
Groetjes,

Anton
  _____________ _______________________________________________________
|             |                                                       |
|  _   _  ___,| K. Anton Feenstra                                     |
| / \ / \'| | | Dept. of Pharmacochem. - Vrije Universiteit Amsterdam |
|(   |   )| | | De Boelelaan 1083 - 1081 HV Amsterdam - Netherlands   |
| \_/ \_/ | | | Tel: +31 20 44 47608 - Fax: +31 20 44 47610           |
|             | Feenstra at chem.vu.nl - www.chem.vu.nl/~feenstra/       |
|             | "If You See Me Getting High, Knock Me Down"           |
|             | (Red Hot Chili Peppers)                               |
|_____________|_______________________________________________________|