Ampere vs Turing architecture in HOOMD

[Carlos Andrés]

Hello everyone.

In our lab, we want to buy new custom workstations to use with HOOMD, and for the GPU we are looking for either Nvidia Ampere GPUs (A2000, A4000) or  Nvidia Turing GPUs (T1000). So our question is: Are there any compatibility issues with HOOMD using Ampere or Turing architectures? Or HOOMD doesn't care about this?

I really appreciate any help you can provide.

Carlos.

[Joshua Anderson]

Carlos,

HOOMD-blue supports all NVIDIA GPUs with compute capability 6.0 and greater. However, It will perform many times faster on GPUS with more FP64 ALUs GPUs such as H100, A100, V100. So many times faster that a good CPU can outperform a FP32 optimized GPU in most benchmarks (your results may vary).

You can find FP64 performance specification of GPUS on sites like techpowerup. I highly recommend using institutional and national HPC resources for production jobs. NVIDIA currently produces no actively cooled workstation cards with high FP64 performance.

------

Joshua A. Anderson, Ph.D.

Research Area Specialist, Chemical Engineering, University of Michigan

--
You received this message because you are subscribed to the Google Groups "hoomd-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hoomd-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hoomd-users/07743460-2177-4dac-863c-e1e706c82870n%40googlegroups.com.