We are a distributed cloud provider (like Airbnb for GPUs). Not sure if I can mention the name here.
We’ve got 400,000+ RTX/GTX consumer GPUs on our network (10k+ active on any day). Recently, we’ve had a few companies run massive molecular dynamics simulations on our network.
Following this, we’d like to publish some benchmarks that would be helpful for new users to evaluate our infra. Here are some questions:
- When selecting a cloud provider for your MD simulations, are there any common benchmarks you look for?
- If so, what would be a good benchmark to replicate on our infra?
- What are some other criteria you look for when selecting a cloud provider for GROMACS?
If you’re open to it, I’d love to have a 15 minute chat for some follow-up questions with folks in the industry (happy to pay for your time).