A time-series database stores application metrics indexed by time, making temporal retrieval efficient for monitoring, dashboards, and alerts. section 3.5
In a Prometheus-based monitoring pipeline, the database:
- Receives metrics scraped by a worker from an application endpoint such as /metrics. section 3.5
- Supplies queried measurements to Grafana for charts and dashboards. section 3.5
- Supports alert evaluation for conditions such as downtime, high response latency, or falling request counts. section 3.5
Prometheus and Grafana
A FastAPI service exposes /metrics. Prometheus scrapes and stores its measurements, a query aggregates request increases over ten minutes by response code, and Grafana renders the result as a time-series chart. section 3.5
Figure 3.17 Application 1 and Application 2 expose an endpoint for metrics. Prometheus scrapes those metrics, which are then visualized by Grafana. Alertmanager uses those metrics to check if any alerts need to be routed to a channel; in this case, it’s an email.
Trace the production flow from application metrics endpoints through Prometheus storage to Grafana visualization and Alertmanager routing. section 3.5
