Grafana is the visualisation layer that sits above other systems rather than a monitoring stack of its own. It queries Prometheus, Loki, PostgreSQL, ClickHouse, and dozens of other sources, and puts them on the same dashboard.
Alerting is unified across those sources, which is what keeps it in the middle of most observability setups even as the data stores underneath change.
Decisive facts
- Dozens of data sources on a single dashboard
- Unified alerting with routing to the usual destinations
- Templated dashboards driven by variables
- Annotations tying deploys and incidents to the timeline
- Self-hostable with SSO and team-scoped permissions