Is it possible to allocate multiple GPUs, such as assigning two GPUs to the frontend and two GPUs to the backend? If so, could you explain how to configure it?
I tried using two GPUs in the backend on Docker, but it seems that only one GPU is being utilized. I’m also wondering if using both GPUs is supported.
In addition, Azure VM uses two separate nodes for the frontend and backend. I would like to get some tips on whether it is possible to use separate nodes for two personal local servers as well.
Multi-GPU is not supported yet, but it is on our road map.
Can you please elaborate on “whether it is possible to use separate nodes for two personal local servers as well.”? In general, yes, you can deploy an on-prem solution with two seperate nodes. We can provide guidance on it.
I am in the process of installing a Dell R750 with 2 RTX 6000 ADA GPUs, and I plan to operate in a colocated setup without connecting nodes. I look forward to Multi-GPU support being enabled on the backend in the future. Thank you!