Can I use 2 Tesla V100 GPUs on, each on a different monitor in a single remote session?

We are using Azure NC12s_v3 VMs that have 2 Tesla V100 GPUs.We would like to connect to such a VM and have 2 monitors - each dedicated to a different GPU. Is this possible?

The reason for that is that we would like to have two instances of TouchDesigner, each working on a different GPU. TouchDesigner GPU affinity is driven by display - so if your display 1 is assigned to GPU 1, this is where the processing will take place.


Unless the application can run multiple instances that allow you to specify which resources each one uses, what you’re asking for won’t work.

Why not simply use 2 Azure VMs with 1 V100 on each and connect a session to each VM?