NVIDIA vGPU allows vSphere to share NVIDIA GPUs among multiple VMs by using either the timesliced vGPU profile or the MIG-with-vGPU profile (we will call this MIG vGPU throughout this paper). These two vGPU modes provide a flexible choice on how GPUs are shared to best leverage the GPU resource.
With two options available, you might wonder if you should choose vGPU or MIG vGPU. This paper explores multiple use cases of the two profiles with different workloads, from ML training to ML inference and other workloads. We also show you how to choose the right profile for your workloads to maximize the benefits of vGPU and MIG vGPU.