Monitoring and Alerts
Cloudera Machine Learning leverages CDP Monitoring based on Prometheus and Grafana to provide dashboards that allow you to monitor how CPU, memory, storage, and other resources are being consumed by your ML workspaces.
Prometheus is an internal data source that is auto-populated with resource consumption data for each deployment. Grafana is the monitoring dashboard that allows you to create visualizations for resource consumption data from Prometheus. By default, CML provides three Grafana dashboards: K8 Cluster, K8s Containers, and K8s Node. You can extend these dashboards or create more panels for other metrics. For more information, see the Grafana documentation.