Observability Wiki
Open source

Grafana

The open-source dashboarding and visualization platform that has become the default front end for metrics, logs, traces, and profiles alike.

Last updated

What it is

Grafana is an open-source platform for querying, visualizing, alerting on, and exploring metrics, logs, and traces regardless of where they’re stored. It was created by Torkel Ödegaard in 2014 as a fork of Kibana focused on time-series dashboards, and is now developed by Grafana Labs, the company Ödegaard co-founded around it. Grafana itself doesn’t store data — it connects to dozens of data sources (Prometheus, Elasticsearch, InfluxDB, CloudWatch, and many more via plugins) and renders dashboards, so it’s typically the visualization layer sitting on top of whatever backends a team already runs.

Grafana Labs also builds and maintains several of the backends commonly paired with it — Grafana Loki for logs, Grafana Tempo for traces, Grafana Mimir for long-term Prometheus-compatible metrics storage, and Grafana Pyroscope for continuous profiling. Together these are informally known as the “LGTM” (or now “LGTM+P”) stack. All of these components, including Grafana itself, are open source, and Grafana Labs sells Grafana Cloud, a fully managed SaaS version of the entire stack, as its commercial offering.

Why teams choose it

  • Vendor-neutral by design. Because Grafana queries external data sources rather than owning storage, teams can swap or mix backends (self-hosted Prometheus, a cloud vendor’s metrics API, Elasticsearch for logs) without changing the dashboarding layer or retraining users.
  • Deep plugin and panel ecosystem. Community and commercial data source plugins cover most databases and cloud APIs, though quality and maintenance vary a lot across the plugin catalog — some are official, others community-maintained and less polished.
  • Alerting is unified but not always simple. Grafana’s built-in alerting can evaluate rules against any connected data source, which is convenient, but managing alert routing and notification policies at scale has a steeper learning curve than a purpose-built alerting tool.
  • The “LGTM” stack is genuinely well-integrated — trace-to-logs and metrics-to-traces correlation works smoothly when using Grafana Labs’ own backends, less so when mixing in third-party stores.

Pricing model

Grafana the visualization tool is free and open-source (AGPLv3-licensed core, with some enterprise-only plugins and features under a commercial license). Grafana Labs’ commercial revenue comes primarily from Grafana Cloud, a hosted SaaS offering with a free tier and paid tiers that scale with usage (active series, log/trace ingestion volume, and similar consumption metrics), plus Grafana Enterprise for self-managed deployments needing enterprise support and extra features.

Alternatives

For dashboarding specifically, Kibana (Elastic’s visualization layer, often paired with Elasticsearch) and Splunk’s native dashboards are the closest open-source-adjacent alternatives. For teams wanting an integrated commercial platform instead of assembling their own stack, Datadog, New Relic, and Dynatrace bundle visualization with their own data collection and storage.

Visit official site →