Infra & metrics

Prometheus

open source / community (CNCF)

The de-facto standard pull-based metrics engine and time-series database for cloud-native and Kubernetes monitoring; everything else in the space orbits it.

Category
Infra & metrics
License
Open source
Deployment
Self-hosted
Cost
Free
Free tier
Yes
Self-host effort
Heavy
Maturity
Incumbent
Popularity
≈64k GitHub stars; category center of gravity

The catch

Single-node by design — no native HA or long-term storage — so any serious deployment becomes a 4-5 component stack (Alertmanager, Grafana, Thanos/Mimir, exporters) you assemble and operate yourself.

Monitors

MetricsServersK8sCloudNetwork

Protocols

PrometheusSNMP

Capabilities

AlertingAutodiscoveryAPIConfig as code

Built for

SRE / DevOpsEnterpriseHomelab

The honest take

The cloud-native default — and the wrong tool to call “a product.” You’re adopting a stack: Prometheus + Alertmanager + Grafana, plus Thanos or Mimir the moment you need HA or history, plus an exporter per thing. For traditional SNMP networks, snmp_exporter works but it’s the clunky path; Zabbix is usually the gentler escape from SolarWinds.

Pricing in the real world

  • Software $0
  • Grafana Enterprise support optional $10-30k/yr

Free software; cost is engineering time + the surrounding stack.