[gmx-users] compiling GROMACS with CUDA shared Libraries

Mark Abraham mark.j.abraham at gmail.com
Thu Nov 10 06:00:46 CET 2016


This will just happen by default, particularly if cmake can't find a static

But you will likely find that actually running on a remote GPU will be
disastrous for performance. GROMACS is engineered around making use of
locality and both the CPU and GPU, in contrast with many of the targets for
things like rCUDA.


On Thu, 10 Nov 2016 04:40 Rizki Bayu <rizki.bayu.rb2 at gmail.com> wrote:

> i currently using GROMACS to test a rCUDA server instance on my laboratory.
> It specified that any programs using rCUDA must be compiled with CUDA
> shared library by specifying it during the compilation (-cudart=shared on
> NVCC, -lcudart on GCC/++). Where do i have to put those option for GROMACS
> compilation?
> Thank you.
> Rizki Bayu
> --
> 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.

More information about the gromacs.org_gmx-users mailing list