[gmx-developers] EM with gmx4.0.2

[Berk Hess]

Hi,

Something strange happened in mdrun, since it says the grid size is 0.
Could you mail me the tpr file that generated this error?

You could use -npme 0, but your relative pme load (0.73) is extremely high.
For better performance you should increase your cut-off and PME grid spacing
by the same factor, grompp probably also gave you a note about this.
With a better relative pme load, mdrun will run fine without further 
options.

Berk


andrea spitaleri wrote:
> Hi there,
> we having trouble in EM using gmx-4.0.2. Basically the problem is that the run crash immediately
> complaining about the number of separate PME:
>
>   
>> mdrun -s box -v -deffnm boxem
>>     
>
> Program mdrun, VERSION 4.0.2
> Source code file: domdec_setup.c, line: 132
>
> Fatal error:
> Could not find an appropriate number of separate PME nodes. i.e. >= 0.727885*#nodes (11) and <=
> #nodes/2 (8) and reasonable performance wis
> e (grid_x=88, grid_y=88).
> Use the -npme option of mdrun or change the number of processors or the PME grid dimensions, see the
> manual for details.
>
> while:
>
>   
>> mdrun -s box -npme 8 -v -deffnm boxem
>>     
> gives:
> Program mdrun, VERSION 4.0.2
> Source code file: domdec.c, line: 5860
>
> Fatal error:
> The size of the domain decomposition grid (0) does not match the number of nodes (8). The total
> number of nodes is 16
>
> We are able to minimize the system using gmx-3.3.3 and then run the MD with gmx-4.0.2 without troubles.
> our em.mdp is:
>
> define              =  -DFLEXIBLE
> integrator          =  steep
> dt                  =  0.002    ; ps
> nsteps              =  2000
> nstlist             =  5
> ns_type             =  grid
> pbc                 =  xyz
> rlist               =  0.9
> table-extension     =  2
> coulombtype         =  PME
> rcoulomb            =  0.9
> rvdw                =  0.9
> fourierspacing      =  0.15
> fourier_nx          =  0
> fourier_ny          =  0
> fourier_nz          =  0
> pme_order           =  4
> ewald_rtol          =  1e-5
> optimize_fft              =  yes
> ;
> ;       Energy minimizing stuff
> ;
> emtol               =  1000
> emstep              =  0.01
> lincs_iter          = 4
>
>
> any help is well accepted.
>
> thanks in advance
>
> andrea
>
>