[gmx-users] Nicelevel default 0 - mdrun_mpi of Gromacs 4.0.2

[João M. Damas]

Hi Berk,

Thanks for your patience. I believe the variable you're talking about
corresponds to LAMWORLD in lam/mpi. Even though I wasn't able to make it
work, I appreciate your help.

Regards,
João

On Fri, Feb 6, 2009 at 6:01 PM, Berk Hess <gmx3 at hotmail.com> wrote:

>  Hi,
>
> VAR should be an environment variable that is set by your MPI
> implementation.
> Try:
> mpirun -np 2 env
>
> For OpenMPI I use:
> OMPI_MCA_universe
>
> Berk
>
> ------------------------------
> Date: Fri, 6 Feb 2009 16:38:57 +0000
>
> Subject: Re: [gmx-users] Nicelevel default 0 - mdrun_mpi of Gromacs 4.0.2
> From: jmdamas at itqb.unl.pt
> To: gmx-users at gromacs.org
>
>
>
> On Fri, Feb 6, 2009 at 4:12 PM, Berk Hess <gmx3 at hotmail.com> wrote:
>
>  Hi,
>
> I always compile Gromacs with --enable_mpi --enable-mpi-environment=...
> such that you can run a serial mdrun without using mpirun.
>
>
> Interesting. So what does that "..." (or VAR in the help of configure)
> means? I'm sorry, but I can't find any information about it anywhere. If
> it's just a variable we come up with, how or when do we set it (and what do
> we set it to) when running the MPI mdrun? It doesn't seem to work when I
> don't set it at all...
>
> Regards,
> João
>
>
>
> Berk
>
> ------------------------------
> Date: Fri, 6 Feb 2009 15:22:54 +0000
> Subject: Re: [gmx-users] Nicelevel default 0 - mdrun_mpi of Gromacs 4.0.2
>
> From: jmdamas at itqb.unl.pt
> To: gmx-users at gromacs.org
>
> On Mon, Jan 19, 2009 at 6:09 PM, Berk Hess <gmx3 at hotmail.com> wrote:
>
>  Hi,
>
> When running in parallel on a cluster through a queuing system
> one usually does not (or can not) see or check the nice level
> of the mdrun processes. But the nice level can have a significant
> impact on performance, especially when running in parallel,
> since all processes have to wait for the slowest one.
> Therefore I set the nicelevel to 0 when running in parallel.
>
> Note that if you configure mdrun with --enable_mpi
> --enable-mpi-environment,
> running an MPI mdrun binary without mpirun will still give nice 19.
>
>
> Hi Berk,
>
> Thanks for the answer and for the tip. It worked out fine. But still, I was
> intrigued when you said "running an MPI mdrun binary without mpirun". What
> do you mean with that? Do you mean that you don't need to launch the mpirun
> before the MPI mdrun?
>
> Regards,
> João
>
>
>
> Berk
>
> ------------------------------
> Date: Mon, 19 Jan 2009 16:29:16 +0000
> From: jmdamas at itqb.unl.pt
> To: gmx-users at gromacs.org
> Subject: [gmx-users] Nicelevel default 0 - mdrun_mpi of Gromacs 4.0.2
>
>
> Hello,
>
> I would like to know if there is any reason for mdrun_mpi to have a
> nicelevel default of zero in Gromacs 4.0.2 instead of the default 19 of
> mdrun of every version or mdrun_mpi of the other versions. I searched the
> mailing list for any discussion related to this but I haven't found
> anything.
>
> Thanks in advance,
> João
>
> --
> João M. Damas
> PhD Student
> Protein Modelling Group
> ITQB-UNL, Oeiras, Portugal
> Tel:+351-214469613
>
> ------------------------------
> Express yourself instantly with MSN Messenger! MSN Messenger<http://clk.atdmt.com/AVE/go/onm00200471ave/direct/01/>
>
> _______________________________________________
> gmx-users mailing list    gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at http://www.gromacs.org/search before posting!
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-request at gromacs.org.
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>
>
>
>
> --
> João M. Damas
> PhD Student
> Protein Modelling Group
> ITQB-UNL, Oeiras, Portugal
> Tel:+351-214469613
>
> ------------------------------
> Express yourself instantly with MSN Messenger! MSN Messenger<http://clk.atdmt.com/AVE/go/onm00200471ave/direct/01/>
>
> _______________________________________________
> gmx-users mailing list    gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at http://www.gromacs.org/search before posting!
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-request at gromacs.org.
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>
>
>
>
> --
> João M. Damas
> PhD Student
> Protein Modelling Group
> ITQB-UNL, Oeiras, Portugal
> Tel:+351-214469613
>
> ------------------------------
> See all the ways you can stay connected to friends and family<http://www.microsoft.com/windows/windowslive/default.aspx>
>
> _______________________________________________
> gmx-users mailing list    gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at http://www.gromacs.org/search before posting!
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-request at gromacs.org.
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>



-- 
João M. Damas
PhD Student
Protein Modelling Group
ITQB-UNL, Oeiras, Portugal
Tel:+351-214469613
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20090209/d21f7a2f/attachment.html>