[gmx-users] cutoff-sch​eme in CPU& GPU

[leila salimi]

Dear GROMACS Users,

I have a question regarding the Neighbor searching, I used group scheme
when I used 4.5 and now I found that there is verlet in version 4.6. When I
compare the trajectories, For verlet simulation I found that the atoms of
water molecules are separated and it caused the long bond in vmd. I mean
that one hydrogen is located on the right side of the cell and the oxygen
and another hydrgon are on the left side that the first hydrogen can come
with periodic image, but these long bond (artifact bonds) are not
convenient and it is difficult for visualization.

I did not have this problem with previous method (group). Could you explain
me what it is happened in  Cut-off schemes = verlet?
And is it possible to have the trajectory with the right bonds (not broken
bond)?

Kind Regards,


On Sun, May 25, 2014 at 5:39 PM, Szilárd Páll <pall.szilard at gmail.com>
wrote:

> On Sat, May 24, 2014 at 9:39 PM, Que Pasa <quepasa41 at gmail.com> wrote:
> > Yes. You can. The entire program was built and was functional on CPUs
> well
> > before GPUs came into the picture. Due to the way calculations are
> > off-loaded to GPUs not all features available on CPUs are available in
> the
> > GPU implementation. "cutoff-scheme = Verlet" is one such example. For
> more
>
> I guess you meant "cutoff-scheme = Group". The Verlet scheme is the
> one that supports GPU acceleration while the (traditionally default up
> until v5.0) Group scheme does not.
>
> Also note that in 5.0 nearly all features will be supported in the
> Verlet scheme!
>
> --
> Szilárd
>
> > details please read http://www.gromacs.org/GPU_acceleration and the
> > relevant papers.
> >
> > hth
> >
> >
> > On Sat, May 24, 2014 at 3:52 AM, Yorquant Wang <wangykoo at gmail.com>
> wrote:
> >
> >> Hi, all
> >>
> >>     Recelly I am testing the speed of GMX4.6 on our new supercomputer.
> >> There are two parameters "cutoff-scheme " and "nstlist ".
> >>        for GPUs I uesd:
> >>            cutoff-scheme=Verlet
> >>            nstlist                  = 20
> >>        I found the speed is fast.
> >>
> >>   I have tested if I use "cutoff-scheme=Verlet" combined with
> >> "nstlist=20" for CPUs, the speed was pretty fast. Could I use the same
> >> parameter for CPUs?
> >>
> >> --
> >> Yukun Wang
> >> PhD candidate
> >> Institute of Natural Sciences && College of Life Science, Shanghai Jiao
> >> Tong University
> >> Cell phone: 13621806236.
> >> China Shanghai
> >> --
> >> 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.
>