TimescaleDB is a PostgreSQL extension that turns Postgres into a time-series database. It adds hypertables, which transparently partition a regular table into chunks by time (and optionally by space), so inserts and range queries stay fast as the table grows into billions of rows.
Because it is an extension, TimescaleDB inherits the full SQL surface, indexes, joins, foreign keys, and tooling of Postgres. On top of that it adds time-series specific features: continuous aggregates that incrementally materialize rollups, compression that converts row-oriented chunks into columnar segments, data retention policies, and hyperfunctions for things like time_bucket, gap-filling, and last-observation-carried-forward.
TimescaleDB is a common backend for metrics, IoT telemetry, financial tick data, and application observability, where teams want the operational familiarity of Postgres without running a separate specialized TSDB like InfluxDB or Prometheus for long-term storage. The company behind it rebranded from Timescale to Tiger Data in 2025, but the open-source extension and repository still use the TimescaleDB name.