Prometheus metrics monitoring
The default metrics engine for containers and Kubernetes, self-hosted, with your data staying on your own infrastructure.
What this covers
Prometheus is an open-source time-series monitoring system and the second project ever to graduate from the Cloud Native Computing Foundation, after Kubernetes. It works by pulling: it scrapes HTTP endpoints exposed by your applications, or by exporters that translate something else into metrics. Queries are written in PromQL, and alert routing, grouping and silencing is handled by Alertmanager alongside it.
Prometheus 3.0 arrived in November 2024, the first major release in seven years, bringing UTF-8 support by default, Remote Write 2.0, native histograms, OTLP ingestion and a rewritten interface. It also carried breaking changes to feature flags, configuration files, PromQL and scrape protocols, so moving from 2.x is a genuine migration rather than a version bump. We handle that upgrade as its own piece of work.
One thing we will tell you up front: Prometheus is not built to be your long-term historical store. If you need a year of retention it wants a companion such as Thanos or Grafana Mimir, and that is a decision worth making at design time rather than discovering at month nine.
What’s included
- Prometheus deployment, on Kubernetes or on plain servers
- Exporter selection and configuration across your estate
- Service discovery and scrape configuration
- PromQL recording and alerting rules
- Alertmanager routing, grouping and silencing
- Migration from Prometheus 2.x to 3.x
- Long-term storage design where retention demands it
Platforms and channels
Need Prometheus metrics monitoring?
Tell us which services matter most and which channels your customers reach them on, and our engineers will come back with a monitoring proposal.