TimescaleDB
TimescaleDB is a PostgreSQL extension that adds automatic time-based partitioning, hybrid row-columnar storage, and incrementally refreshed aggregates, keeping full PostgreSQL SQL and its entire ecosystem of drivers and tools.
LicenseSource-available
DeploymentSelf-hostedManaged
PricingFreeSubscription
Operational complexityLow
WorkloadInteractiveStreaming
LanguageC
Use it when
- You already run PostgreSQL and want time-series scale without a second system: the same drivers, ORMs, backups, and SQL.
- Recent data should stay mutable rows while history compresses into columns (Hypercore) inside the same table.
- Rollups should maintain themselves; continuous aggregates refresh incrementally in the background.
- Full UPDATE and DELETE, and joins between time-series and business tables, are requirements.
Think twice when
- You need scale-out; multi-node support was removed in 2.14, so the ceiling is one server plus read replicas and object-storage tiering in Tiger Cloud.
- You would offer it as a service; the Timescale License forbids database-as-a-service use of the tsl-licensed features, where several headline capabilities live.
- Peak ingest and scan rates matter most; purpose-built engines like ClickHouse and QuestDB exceed what a Postgres extension reaches.
How it runs
Install the extension into PostgreSQL: hypertables partition by time automatically, background workers compress chunks and refresh continuous aggregates, and everything connects as ordinary Postgres. Tiger Cloud (the company renamed from Timescale to TigerData) runs it managed with pooling, replicas, and S3 tiering.
Details
CompareHow TimescaleDB answers the questions OLAP Engines turns on.
OLAP Engines
| How it works | |
| Specialization | PostgreSQL extension: time-series hypertables on a Postgres server |
| SQL dialect | PostgreSQL |
| Joins | Full PostgreSQL joins |
| Updates and deletes | Full PostgreSQL UPDATE and DELETE, columnstore chunks included |
| Ingest to queryable | Immediate on commit; rows land in the rowstore first |
| Storage format | PostgreSQL storage; Hypercore row and columnar |
| Compute and storage | Single node: multi-node removed in 2.14; Tiger Cloud tiers to S3 |
| Indexes | PostgreSQL indexes; automatic minmax and bloom sparse on columnstore |
| Pre-aggregation | Continuous aggregates: incremental, stackable, real-time |
| Running it | |
| Concurrent users | PostgreSQL connections; pooling and read replicas in Tiger Cloud |
Alternatives to TimescaleDB
