[gmx-developers] Parallelism in trajectory analysis tools

[Christoph Junghans]

On Tue, Aug 14, 2018 at 07:25 Kutzner, Carsten <ckutzne at gwdg.de> wrote:

>
>
> > On 14. Aug 2018, at 14:30, Szilárd Páll <pall.szilard at gmail.com> wrote:
> >
> > IIRC there did not seem to be enough use cases for parallel analysis  at
> the time (considering the limitations?), but I'm not sure. Unless I'm
> remembering incorrectly, only independent per-frame processing was
> possible, right?
> So for an analysis that only needs the data of a single frame,
> parallel processing should already work?
>
> At the moment we wrap the analysis tool by a script
> that calls -b begin -e end for N parts, but then
> one ends up with N output files, which is a bit
> inconvenient.

Hehe, we did the same:
https://github.com/votca/multi_g_rdf

A real parallel analyser would be better though!

Christoph

>
>
> Carsten
>
>
> >
> > --
> > Szilárd
> >
> >
> > On Tue, Aug 14, 2018 at 9:37 AM Kutzner, Carsten <ckutzne at gwdg.de>
> wrote:
> > Hi Roland,
> >
> > thanks a lot, this seems like a good starting point for building
> > an MPI-parallel tool! Is there any specific reason why this
> > patch is lying dormant? Any road blocks, or just lack of
> > resources?
> >
> > Best regards,
> >   Carsten
> >
> >
> > > On 14. Aug 2018, at 01:22, Schulz, Roland <roland.schulz at intel.com>
> wrote:
> > >
> > > Hi,
> > >
> > > We done some work on this at UT/ORNL in 2012/13. You find this at
> > > https://gerrit.gromacs.org/c/1316/ . As far as I know no one has done
> anything since.
> > >
> > > Roland
> > >
> > >> -----Original Message-----
> > >> From: gromacs.org_gmx-developers-bounces at maillist.sys.kth.se
> > >> [mailto:gromacs.org_gmx-developers-bounces at maillist.sys.kth.se] On
> > >> Behalf Of Kutzner, Carsten
> > >> Sent: Thursday, August 9, 2018 7:46 AM
> > >> To: <gmx-developers at gromacs.org> GROMACS development <gmx-
> > >> developers at gromacs.org>
> > >> Subject: [gmx-developers] Parallelism in trajectory analysis tools
> > >>
> > >> Hi,
> > >>
> > >> what is the status of using any type of parallelism to speed up
> trajectory
> > >> analysis tools?
> > >>
> > >> I guess OpenMP should be no problem, but what about MPI?
> > >>
> > >> What are the plans here / what would be preferred in future patches?
> > >>
> > >> Thanks!
> > >>  Carsten
> > >>
> > >> --
> > >> Gromacs Developers mailing list
> > >>
> > >> * Please search the archive at
> > >> http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_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-developers or
> > >> send a mail to gmx-developers-request at gromacs.org.
> > > --
> > > Gromacs Developers mailing list
> > >
> > > * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_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-developers
> or send a mail to gmx-developers-request at gromacs.org.
> >
> > --
> > Dr. Carsten Kutzner
> > Max Planck Institute for Biophysical Chemistry
> > Theoretical and Computational Biophysics
> > Am Fassberg 11, 37077 Goettingen, Germany
> > Tel. +49-551-2012313, Fax: +49-551-2012302
> > http://www.mpibpc.mpg.de/grubmueller/kutzner
> > http://www.mpibpc.mpg.de/grubmueller/sppexa
> >
> > --
> > Gromacs Developers mailing list
> >
> > * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_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-developers
> or send a mail to gmx-developers-request at gromacs.org.
> > --
> > Gromacs Developers mailing list
> >
> > * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_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-developers
> or send a mail to gmx-developers-request at gromacs.org.
>
> --
> Dr. Carsten Kutzner
> Max Planck Institute for Biophysical Chemistry
> Theoretical and Computational Biophysics
> Am Fassberg 11, 37077 Goettingen, Germany
> Tel. +49-551-2012313, Fax: +49-551-2012302
> http://www.mpibpc.mpg.de/grubmueller/kutzner
> http://www.mpibpc.mpg.de/grubmueller/sppexa
>
> --
> Gromacs Developers mailing list
>
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_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-developers
> or send a mail to gmx-developers-request at gromacs.org.

-- 
Christoph Junghans
Web: http://www.compphys.de
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-developers/attachments/20180814/512aa98d/attachment-0001.html>