I've been assigned a job which consists of deploying a NVIDIA vGPU solution on an Microsoft Hyper-V hypervisor.
I've understood the concept pretty well but there is now this thing that is bothering me.
I am supposed to virtualize a single physical GPU into x instances and from what I've read, NVIDIA vGPU Manager is the product I'm looking for.
Hyper-V is only quoted in "pass-through" configurations in the official documentation but isn't ever told to be unsuitable for the vGPU Manager.
I'm now confused if it is possible to deploy such a thing !
question from:
https://stackoverflow.com/questions/65951263/nvidia-vgpu-manager-on-microsoft-hyper-v 与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…