[gmx-users] GPU performance, regarding

Mateusz Bieniek bieniekmat at gmail.com
Thu Dec 12 14:35:09 CET 2019


Hi Gromacs,

A small digression: Ideally Gromacs would make it more clear in the error
message explaining which part is not implemented for the GPUs.

Thanks, Mat

On Thu, 12 Dec 2019 at 13:01, RAHUL SURESH <drrahulsuresh at gmail.com> wrote:

> Hi John
>
> Thank you and adding here the mdp settings
>
> title       = Protein-gas molecule interaction MD simulation
> ; Run parameters
> integrator  = md
> nsteps      = 200000000
> dt          = 0.002     ; 2 fs
> ; Output control
> nstxout     = 5000        ;
> nstvout     = 5000  ;
> nstenergy   = 5000    ;
> nstlog      = 5000      ;
> energygrps  = Protein
> ; Bond parameters
> continuation    = yes           ;
> constraint_algorithm = lincs
> constraints     = all-bonds     ;
> lincs_iter      = 1             ;
> lincs_order     = 4
> ; Neighborsearching
> cutoff-scheme = Verlet
> ns_type     = grid      ;
> nstlist     = 20         ; 10 fs
> rlist       = 1.4       ;
> rcoulomb    = 1.4       ;
> rvdw        = 1.4       ;
> ; Electrostatics
> coulombtype     = PME       ;
> pme_order       = 4         ;
> fourierspacing  = 0.16      ;
> ; Temperature coupling
> tcoupl      = V-rescale                     ;
> tc-grps     = Protein_NO2 Water_and_ions ;
> tau_t       = 0.1   0.1                     ;
> ref_t       = 300   300                     ;
> ; Pressure coupling
> pcoupl      = Parrinello-Rahman             ;
> pcoupltype  = isotropic                     ;
> tau_p       = 2.0                           ;
> ref_p       = 1.0                           ;
> compressibility = 4.5e-5
> ; Periodic boundary conditions
> pbc         = xyz       ; 3-D PBC
> ; Dispersion correction
> DispCorr    = EnerPres  ;
> ; Velocity generation
> gen_vel     = no
>
> On Thu, Dec 12, 2019 at 6:01 PM John Whittaker <
> johnwhittake at zedat.fu-berlin.de> wrote:
>
> > Hi,
> >
> > > Hi Users.
> > >
> > > I am simulating a peptide of 40 residues with small molecules using
> > oplsaa
> > > ff in Gromacs 2018.20 installed in CUDA environment.. The workstation
> has
> > > 16 Cores and 2 1080Ti card On execution of command gmx_mpi mdrun -v
> > > -deffnm
> > > md for 100ns it shows no usage of gpu card. For the command  gmx_mpi
> > mdrun
> > > -v -deffnm md -gputasks 01 -nb gpu, the job is terminated with note
> > >  " NB interaction on the gpu were required but not supported for these
> > > simulation settings. change your settings or do not require using
> gpus."
> >
> > You should provide the content of your .mdp file. According to the error
> > message, some of your settings are not compatible with GPU acceleration.
> > Without some information about your settings, no one can really help.
> >
> > - John
> >
> > >
> > > Could anyone explain a solution on this issue?
> > >
> > > Thank you
> > >
> > > --
> > > *Regards,*
> > > *Rahul *
> > > --
> > > 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.
> > >
> >
> >
> > --
> > 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.
> >
>
>
> --
> *Regards,*
> *Rahul *
> --
> 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.
>


More information about the gromacs.org_gmx-users mailing list