Viewing the Resource Usage of a Training Job¶
Operations¶
You can view the resource usage of a compute node in the Resource Usages window. The data of at most the last three days can be displayed. When the resource usage window is opened, the data is loading and refreshed periodically.
Operation 1: If a training job uses multiple compute nodes, choose a node from the drop-down list box to view its metrics.
Operation 2: Click cpuUsage, gpuMemUsage, gpuUtil, or memUsage to show or hide the usage chart of the parameter.
Operation 3: Hover the cursor on the graph to view the usage at the specific time.
Parameter | Description |
|---|---|
cpuUsage | CPU usage |
gpuMemUsage | GPU memory usage |
gpuUtil | GPU usage |
memUsage | Memory usage |