The dashboards behind your dashboards

If you've ever seen a wall of glowing graphs monitoring a website's health, there's a good chance you were looking at Grafana. It's an open-source platform (76,000+ stars on GitHub) for turning raw numbers — CPU usage, request counts, error rates — into dashboards you can actually read at a glance.

Grafana's key trick is that it doesn't store any data itself. Instead it connects to data sources like Prometheus (metrics), Loki (logs), Elasticsearch, or plain Postgres, queries them live, and renders the results. This separation means one dashboard can pull from many systems at once, and you're never locked into a single database.

The three pillars it visualizes are often called metrics, logs, and traces — the core signals of observability, the practice of understanding what a running system is doing from the outside. When something breaks at 3 a.m., an on-call engineer usually reaches for a Grafana dashboard first to spot where the graph went wrong.

A surprising bit of history: Grafana began in 2014 as a fork of Kibana, the visualization tool tied to Elasticsearch. Its creator wanted the same slick graphs but for time-series metrics — and that fork grew into a company and an ecosystem used across the industry.