Use the Private AI (GPU) dashboards to monitor and troubleshoot GPU issues in VMware Aria Operations.

To access the dashboards, from the left menu, click Visualize > Dashboards. From the Dashboards panel, navigate to All > Private AI.

The following dashboards are available:

Dashboard Name Purpose
GPU Equipped Clusters Use this dashboard for the details related to GPU compute utilization and GPU memory usage at the cluster level, host level, and GPU level.
GPU Overview Use this dashboard to view if any GPUs have high-temperature. This dashboard also highlights GPUs with low to zero usage by analyzing their capacity based on compute and memory utilization.