Prometheus
Time-series metrics collection and alerting engine — the Kubernetes-native monitoring standard.
Self-hosted Open source Apache 2.0 metricsalertingkubernetes
Pricing
Free (open source).
Prometheus pulls metrics from your services on a scrape interval and stores them as time series, with PromQL for querying and Alertmanager for routing alerts. It’s the de facto standard for Kubernetes environments.
Key features
- Pull-based scraping model, no agent to babysit
- PromQL is powerful once you climb the learning curve
- Massive exporter ecosystem (node, Postgres, Redis, etc.)
- Native Kubernetes service discovery
Self-hosted vs. cloud
Self-hosted by default. Long-term storage at scale typically pairs it with Thanos, Cortex, or Mimir rather than running vanilla Prometheus alone past a certain size.
Quick start
docker run -d -p 9090:9090 -v ./prometheus.yml:/etc/prometheus/prometheus.yml prom/prometheus
Alternatives
Grafana is the natural dashboard layer on top. Datadog replaces the whole stack with a managed SaaS.