We also have NVIDIA Grid cards (M10) and I was wondering if I could monitor them 24/7 
For monitoring we are using PRTG and after lots of testing I figured out an option to monitor the vGPU utilisation.

Monitoring all the NVIDIA GPU's usage can only be done within the nvidia-smi CLI command. So I have made a script, that reads out this GPU usage and generates an output that PRTG could read. Finally in PRTG I used the SSH Script that runs this script and reads out the output.

I posted this solution into the PRTG knowledgebase, and can be found here: https://kb.paessler.com/en/topic/82377-monitor-gpu-usage-with-prtg

I'm aware that not everybody is using PRTG. But they have a free version that can monitor 100 sensors. That are enough sensors for monitoring your NVIDIA vGPU