[gmx-users] benchmark of RIB

[Kutzner, Carsten]

Hi tarzan p,

> Am 02.09.2019 um 15:57 schrieb tarzan p <tarzan11_11 at yahoo.com>:
> 
> Hi allI am sure many of you guys would have run a benchmark of RIB system as given https://www.mpibpc.mpg.de/grubmueller/bench
> 
> 
> I could manage 12ns/day with a workstation with 2 x Gold 6148 (20 cores each) and 2 x V100 GPUs. 
> Can you guys post your performance results on a single node with a similar configuration
This sounds like the performance of a single simulation spread over the whole node.
Note that in the „More Bang …“ paper we report *aggregate* performances for running N
simulations on a node with N GPUs - resulting in about 20 ns/day on your example node.
On that node you can even get 27 ns/day of MD trajectory if you run 4 simulations,
i.e. two per GPU.

Best,
  Carsten


> 
> With best wishes
> -- 
> 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.

--
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