[gmx-users] How to increase the volatile gpu-util

[Mark Abraham]

Hi,

On Fri, 28 Jul 2017 10:37 leila karami <karami.leila1 at gmail.com> wrote:

> Dear gromacs users,
>
> I am running CG MD simulation on a Rocks cluster using the following
> command:
>
> gmx_mpi mdrun -nb gpu -v -deffnm gpu_md -ntomp 16 -gpu_id 0>& md.job &
>
> I tried the different value for the -ntomp option. With -ntomp 16 run
> will finish sooner than the other values for -ntomp.
>
> In this state, volatile gpu-util is 47%. How to increase this parameter?
>

Much more relevant is what gromacs reports in the end of the log file.
Sharing a whole log file via a file sharing service is a useful way to get
some tips.

Mark

nvidia-smi command:
>
>
> |*-------------------------------+----------------------+----------------------+
> *|* GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr.
> *ECC |
> |* Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute
> *M. |
>
> |*===============================+======================+======================|
> *|*   2  TITAN X (Pascal)    On   | 0000:C1:00.0     Off |
> * N/A |
> |* 43%   71C    P2   130W / 250W |    183MiB / 12189MiB |     47%   E.
> *Process |
>
> +-------------------------------+----------------------+----------------------+
>
> Any help will be highly appreciated.
> --
> 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.
>