Ning,
just to add to Pujan's response:
I experienced some speed-up for parallel GPU jobs on our systems with 4x
NVIDIA A100 and NVlink: the fast interconnect is crucial for this speed-up.
But, as Pujan already suggested, it is more efficient to run multiple
jobs on single GPUs, e.g. different trajectories of the same system.
Best,
Anselm
Bioinformatik | NHR.FAU
Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU)
Germany
Am 10.11.2023 um 02:03 schrieb ning via AMBER:
> Thanks for you quick reply. Yes, as you said, I'm trying to make a single long trajectory. Maybe I should to try another enhanced sampling method for exploring more protein conformation.
>
>
> Sincerely,
> Ning
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> At 2023-11-10 08:54:24, "PUJAN AJMERA" <pujanajmera1.g.ucla.edu> wrote:
>
> Hi Ning,
>
>
> Are you aiming to make a single long trajectory faster? Generally, GPUs will not run much faster beyond 1, due to communication overhead (GPUs are much faster than the communication between them), and unsurprisingly in your case, actually slow it down.
>
>
> If you are looking to just get good sampling, I may suggest splitting your trajectory length into 8 runs, and running each on 1 GPU.
>
>
> All the best,
> Pujan Ajmera
> PhD Student at UCLA
>
>
> On Thu, Nov 9, 2023 at 4:46 PM ning via AMBER <amber.ambermd.org> wrote:
>
> Hi amber experts,
>
>
> I have a question about how to use pmemd.cuda.MPI for long time simulation. Because, I found that if applying this command like "mpirun -np 8 pmemd.cuda.MPI .... ” to use 8×2080Ti GPU, the simulation performance had been decreased to 12.5% of one GPU. What's wrong about this ？ I just compiled the parallel version as manual guidance. How to fix this problem? Thanks for any suggestions.
>
>
> Best,
> Ning
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Nov 10 2023 - 01:30:02 PST
