[gmx-developers] implicit solvent simulation with GPUs

Szilárd Páll szilard.pall at cbr.su.se
Mon Nov 19 23:13:01 CET 2012


Hi,

The dhfr benchmark files were prepared for the OpenMM-accelerated mdrun
(called mdrun-gpu in 4.5 and renamed to  mdrun-openmm in 4.6 to avoid
confusion with the native GPU acceleration).

However, the native GPU acceleration does not support implicit solvent
simulation, for those purposes you should build mdrun-openmm (GMX_GPU=ON in
CMake).

You you preferably use the 4.6 code-base (release-4-6 branch) which is more
stable than the 5.0 code-base (master branch) that you are trying to use.

Cheers,

--
Szilárd


On Fri, Nov 16, 2012 at 10:42 PM, Thomas Evangelidis <tevang3 at gmail.com>wrote:

> Dear GROMACS developers,
>
> I am confused about the supporting features of GROMACS-GPU VERSION
> 5.0-dev-20121027-d2934a9-dirty regarding implicit solvent simulations. I
> have downloaded the benchmark set from
> http://www.gromacs.org/Documentation/Installation_Instructions/GPUs#Benchmark_systemsbut cannot run the GPU/dhfr-impl-1nm.bench, GPU/dhfr-impl-2nm.bench and
> GPU/dhfr-impl-inf.bench cases. I.e. with the dhfr-impl-1nm.bench I get:
>
> mdrun_intel_cuda5 -v -s topol.tpr -debug 1
>
> NOTE: GPU(s) found, but the current simulation can not use GPUs
>       To use a GPU, set the mdp option: cutoff-scheme = Verlet
>       (for quick performance testing you can use the -testverlet option)
>
>
> but -testverlet yields:
>
> mdrun_intel_cuda5 -v -s topol.tpr -debug 1 -testverlet
>
> Fatal error:
> Can only convert old tpr files to the Verlet cut-off scheme with 3D pbc
>
> Were these benchmarks made for the older versions of GROMACS-GPU that
> implemented OpenMM? Otherwise how can I run them as they are now?
>
> thanks in advance,
> Thomas
>
>
>
>
>
> --
> gmx-developers mailing list
> gmx-developers at gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-developers
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-developers-request at gromacs.org.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-developers/attachments/20121119/f23f592d/attachment.html>


More information about the gromacs.org_gmx-developers mailing list