Prometheus

Prometheus is used to monitor infrastructure metrics. It scrapes any endpoints that expose metrics in a compatible format. The metrics are stored in Prometheus, and you can query them using PromQL (Prometheus Query Language). Prometheus is not intended for long-term metrics storage, but it supports a variety of remote storage solutions for that purpose.

TimescaleDB can use Prometheus as a remote store for long-term metrics, by using Promscale. Promscale supports both PromQL and SQL queries. PromQL queries are directed to the Promscale endpoint or the Prometheus instance. SQL queries are handled directly by TimescaleDB. Promscale also offers other native time-series capabilities, such as automatically compressing your data, retention policies, continuous aggregate views, downsampling, data gap-filling, and interpolation. Additionally, Promscale supports Grafana using Prometheus and PostgreSQL data sources.