[gmx-users] FEP calculations on multiple nodes

Mark Abraham mark.j.abraham at gmail.com
Thu Aug 24 21:09:57 CEST 2017 

Hi,

It is intended that they work on GPUs, but the implementation of FEP
support for FB restraints pre-dates GPUs for nonbonded interactions, and is
probably not tested, so could be broken.

Mark

On Thu, 24 Aug 2017 18:03 Vikas Dubey <vikasdubey055 at gmail.com> wrote:

> Hi,
> I am also running calculation on GPUs. If FB restraints are only limited to
> CPUs, I am sorry I was not aware of that.
>
> On 24 August 2017 at 17:55, Mark Abraham <mark.j.abraham at gmail.com> wrote:
>
> > Hi,
> >
> > Thanks. That should not be the problem, because all such computations are
> > only on the CPU... But hopefully we will see.
> >
> > Mark
> >
> > On Thu, 24 Aug 2017 17:35 Leandro Bortot <leandro.obt at gmail.com> wrote:
> >
> > > Hello all,
> > >
> > >      This may add something: I had Segmentation Fault using flat-bottom
> > > restraints with GPUs before. I just assumed that this type of restraint
> > was
> > > not supported by GPUs and moved to a CPU-only system.
> > >      Sadly it was some time ago and I don't have the files anymore.
> > >
> > > Best,
> > > Leandro
> > >
> > >
> > > On Thu, Aug 24, 2017 at 5:13 PM, Mark Abraham <
> mark.j.abraham at gmail.com>
> > > wrote:
> > >
> > > > Hi,
> > > >
> > > > Thanks. Good lesson here - try simplifying until things work. That
> does
> > > > suggest there is a bug in flat bottomed position restraints. Can you
> > > please
> > > > upload a tpr with those restraints, along with a report at
> > > > https://redmine.gromacs.org so we can reproduce and hopefully fix
> it?
> > > >
> > > > Mark
> > > >
> > > > On Thu, 24 Aug 2017 16:55 Vikas Dubey <vikasdubey055 at gmail.com>
> wrote:
> > > >
> > > > > Hi,
> > > > >
> > > > > I have just checked with normal restraints. it works fine.
> Simulation
> > > > crash
> > > > > with flat bottom restraints.
> > > > >
> > > > > On 24 August 2017 at 16:43, Mark Abraham <mark.j.abraham at gmail.com
> >
> > > > wrote:
> > > > >
> > > > > > Hi,
> > > > > >
> > > > > > Does it work if you just have the normal position restraints, or
> > just
> > > > > have
> > > > > > the flat-bottom restraints? In particular, I could image the
> latter
> > > are
> > > > > not
> > > > > > widely used and might have a bug.
> > > > > >
> > > > > > Mark
> > > > > >
> > > > > > On Thu, Aug 24, 2017 at 4:36 PM Vikas Dubey <
> > vikasdubey055 at gmail.com
> > > >
> > > > > > wrote:
> > > > > >
> > > > > > > Hi everyone,
> > > > > > >
> > > > > > > I have found out that positions restrains is the issue in my
> FEP
> > > > > > > simulation.  As soon as I switch off position restraints it
> works
> > > > > fine. I
> > > > > > > have the following the restraint file for the ions in my system
> > (I
> > > > > don't
> > > > > > > see any problems with it):
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > *[ position_restraints ]; atom  type      fx      fy      fz
>   1
> > > >  1
> > > > > > 0
> > > > > > > 0  1000     2     1  0  0  1000     3     1  0  0  1000     4
> >  1
> > > > > 0  0
> > > > > > > 1000    5     1  0  0  1000     6     1  0  0  1000     8     1
> > > 0  0
> > > > > > > 1000     9     1  0  0  1000    10     1  0  0  1000    11
> >  1  0
> > > > 0
> > > > > > > 1000    12     1  0  0  1000    13     1  0  0  1000    14
> >  1  0
> > > > 0
> > > > > > > 1000    15     1  0  0  1000    16     1  0  0  1000    17
> >  1  0
> > > > 0
> > > > > > > 1000    18     1  0  0  1000    19     1  0  0  1000    20
> >  1  0
> > > > 0
> > > > > > > 1000    21     1  1000  1000  1000;[ position_restraints ] ;
> flat
> > > > > bottom
> > > > > > > position restraints, here for potassium in site I;  type, g(8
> > for a
> > > > > > > cylinder), r(nm), k    7      2    8      1  1000*
> > > > > > >
> > > > > > >
> > > > > > > On 22 August 2017 at 14:18, Vikas Dubey <
> vikasdubey055 at gmail.com
> > >
> > > > > wrote:
> > > > > > >
> > > > > > > > Hi, I use the following script for my cluster. Also, I think
> > > > problem
> > > > > is
> > > > > > > > calculation specific. I have run a quite a few normal
> > > simulations ,
> > > > > it
> > > > > > > > works fine :
> > > > > > > >
> > > > > > > >
> > > > > > > > #SBATCH --job-name=2_1_0
> > > > > > > > #SBATCH --mail-type=ALL
> > > > > > > > #SBATCH --time=24:00:00
> > > > > > > > #SBATCH --nodes=1
> > > > > > > > #SBATCH --ntasks-per-node=1
> > > > > > > > #SBATCH --ntasks-per-core=2
> > > > > > > > #SBATCH --cpus-per-task=4
> > > > > > > > #SBATCH --constraint=gpu
> > > > > > > > #SBATCH --output out.txt
> > > > > > > > #SBATCH --error  err.txt
> > > > > > > > #========================================
> > > > > > > > # load modules and run simulation
> > > > > > > > module load daint-gpu
> > > > > > > > module load GROMACS
> > > > > > > > export OMP_NUM_THREADS=$SLURM_CPUS_PER_TASK
> > > > > > > > export CRAY_CUDA_MPS=1
> > > > > > > >
> > > > > > > > srun -n $SLURM_NTASKS
> --ntasks-per-node=$SLURM_NTASKS_PER_NODE
> > -c
> > > > > > > > $SLURM_CPUS_PER_TASK gmx_mpi mdrun -deffnm md_0
> > > > > > > >
> > > > > > > > On 22 August 2017 at 06:11, Nikhil Maroli <
> scinikhil at gmail.com
> > >
> > > > > wrote:
> > > > > > > >
> > > > > > > >> Okay, you might need to consider
> > > > > > > >>
> > > > > > > >> gmx mdrun -v -ntmpi XX -ntomp XX -deffnm XXXX  -gpu_id XXX
> > > > > > > >>
> > > > > > > >>
> > > > > > > >>
> > > > > > > >>
> http://manual.gromacs.org/documentation/5.1/user-guide/mdrun
> > > > > > > >> -performance.html
> > > > > > > >>
> > > > > > > >> http://www.gromacs.org/Documentation/Errors#There_is_no_
> > > > > > > >> domain_decomposition_for_n_nodes_that_is_compatible_with_the
> > > > > > > >> _given_box_and_a_minimum_cell_size_of_x_nm
> > > > > > > >> --
> > > > > > > >> 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.
> > > > > > >
> > > > > > --
> > > > > > 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.
> > > > >
> > > > --
> > > > 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.
> > >
> > --
> > 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.
>

More information about the gromacs.org_gmx-users mailing list