Prometheus

A metrics-based monitoring system that pulls time series over HTTP, stores them in a local time series database, and queries them with PromQL.

Category: Observability
CNCF maturity: Graduated
Language: Go
License: Apache-2.0
Repository: prometheus/prometheus
Documented at commit: fc561264 (release-3.13, 3.13.0-rc.0)

What it is

Prometheus collects numeric time series. It discovers targets, scrapes their HTTP metrics endpoints on an interval, and writes the samples to a local time series database. Each series is identified by a metric name plus a set of key/value labels, so the same metric can be sliced by instance, job, or any other dimension. PromQL queries that data for dashboards, alerts, and ad-hoc investigation.

A single Prometheus server is autonomous. It does not depend on distributed storage, and the binary that scrapes is also the binary that stores and serves queries (README:28-33). The multi-dimensional data model and PromQL are the features the project lists first as its distinguishing traits (README:26-35).

It sits at the metrics layer of an observability stack. Alertmanager handles alert routing downstream, exporters expose third-party systems as metrics, and Grafana is the usual visualization front end. For high availability and long-term retention, an external layer such as Thanos or Mimir is stacked on top.

When to use it

You need metrics-based monitoring of infrastructure or services that expose (or can be made to expose) an HTTP metrics endpoint.
You run dynamic infrastructure such as Kubernetes, where service discovery beats static target lists.
You want a query language for ad-hoc analysis and alerting, not just fixed dashboards.
It is a poor fit when you need event logging or distributed tracing; those are separate signals with separate tools.
A single server hits limits on retention and cardinality, since storage is local disk and memory scales with active series. At that point you add a remote layer rather than scaling Prometheus itself.

In this deep-dive

History: origin, milestones, and why it exists.
Architecture: components and how requests flow.
Adoption & Ecosystem: who runs it and what surrounds it.
Internals: the code paths that matter, read from source.
Getting Started: install and a first working setup.

Prometheus ​

What it is ​

When to use it ​

In this deep-dive ​

Sources ​

Prometheus

What it is

When to use it

In this deep-dive

Sources