Hello everyone, I encountered some issues while setting up vGPU.
Currently, my server has one A5000 and one A6000 installed. Both are working well with vGPU and can be properly passed through to virtual machines.
Due to work requirements, we recently added a new RTX 5000 Ada to the same machine. However,
I tried reinstalling the driver, but the installer displayed two different warnings, as shown below:
1.
2.
I would like to ask:
Is it not possible to use the same vGPU driver for GPUs with different architectures (Ampere & Ada) on the same machine?
If so, is there any way to install different drivers for different GPUs to make them work together on the same server?
Thanks for your help!
Currently, my server has one A5000 and one A6000 installed. Both are working well with vGPU and can be properly passed through to virtual machines.
Due to work requirements, we recently added a new RTX 5000 Ada to the same machine. However,
nvidia-smi is unable to recognize this GPU correctly.I tried reinstalling the driver, but the installer displayed two different warnings, as shown below:
1.
WARNING: This system requires the open GPU kernel modules, but contains GPUs which are not supported by the open GPU kernel modules. The unsupported GPUs will be ignored.2.
WARNING: The 'NVIDIA Proprietary' kernel modules are incompatible with the GPU(s) detected on this system. They will be installed anyway, but are not expected to work.I would like to ask:
Is it not possible to use the same vGPU driver for GPUs with different architectures (Ampere & Ada) on the same machine?
If so, is there any way to install different drivers for different GPUs to make them work together on the same server?
Thanks for your help!