Kubernetes Engine support for deploying the NVIDIA GPU Operator cluster add-on

You can now use the NVIDIA GPU Operator add-on to automate the management of NVIDIA software components required to provision GPU nodes. The NVIDIA GPU Operator add-on manages components such as the NVIDIA device plugin, NVIDIA Container Toolkit, MIG Manager, DCGM and DCGM Exporter to enable GPU workloads, monitoring, and GPU sharing capabilities in Kubernetes clusters.

For more information, see Overview of Cluster Add-ons and NVIDIA GPU Operator add-on configuration arguments.