Prometheus
Prometheus is the CNCF metrics database and scraper ecosystem for time-series monitoring and alerting. It pulls metrics from exporters and supports PromQL queries consumed by Grafana and alertmanager. We trial it as the default metrics backend with OTel or native instrumentation.
Blurb
Prometheus is an open-source systems monitoring and alerting toolkit.
Summary
What it is: Pull-based scraper, TSDB storage, PromQL, alert rules, and federation patterns for Kubernetes and services.
When to use: Cloud-native metrics with Grafana dashboards; kube-prometheus-stack style deployments.
When to skip: Long-term log archival (use Loki or Elasticsearch). Vendor SaaS mandate (Datadog).
Key features: Service discovery, recording rules, Alertmanager routing, remote write/read integrations.
Details
| Topic | Notes |
|---|---|
| Deploy | Single binary, Operator on Kubernetes, or managed compatible (Mimir, Cortex, AMP) |
| Pairing | OpenTelemetry metrics pipeline can remote-write into Prometheus-compatible stores |
References