TimescaleDB is an open-source relational database for time-series data. It uses full SQL and is just as easy to use as a traditional relational database, yet scales in ways previously reserved for NoSQL databases.
TimescaleDB uses the same PostgreSQL you know and love, with full SQL, rock‑solid reliability, and the largest ecosystem of developer and management tools.
- Full SQL, powerful analytics, no restrictions
- Leverage your favorite PostgreSQL extensions
- Your entire toolset works: ORMs, connectors, JDBC, applications
- Connect Prometheus for all your observability metrics
- Visualize data using the dashboards you love
Achieve 10-100x faster queries than PostgreSQL, InfluxDB, and MongoDB. Leverage query parallelization, continuous aggregates, and other performance optimizations.
- Achieve 10x faster inserts and ingest 1.5M+ more metrics per second per server for high-cardinality workloads
- Optimized time-series queries and advanced time-series analytics
- Real-time insights over automated continuous aggregations
- Fast scans over compressed columnar storage
- Query faster over longer time horizons with downsampling
Write millions of data points per second. Store hundreds of terabytes on a single node or petabytes across multiple nodes. Handle high‑cardinality data easily.
- Store hundreds of billions of rows and hundreds of TBs of data per server with compression, or tens of TBs uncompressed
- Record billions of distinct time‑series, collect everything you need
- Best‑of‑breed datatype‑specific compression for 16x storage capacity
- Create distributed hypertables across many TimescaleDB nodes
- Parallelize scans and aggregation queries across many nodes
Simplify your stack and store your relational data alongside time‑series data. Ask more complex queries, and build more powerful applications faster.
- Centralize storage of time‑series, application, and sensor data
- Correlate metrics with business and system‑of‑record data
- Unlimited metadata, don't worry about cardinality of labels
- Perform JOINs to understand relations with time‑series data
- Ensure clean, correct data with foreign keys and constraints
Let us run TimescaleDB for you, fully managed on AWS, Azure, or GCP in over 75 regions, with a top-rated support team to ensure your success.
- Spin up a pre‑configured instance in 30 seconds, pay‑as‑you‑go
- Effortless upgrades, fully managed without downtime
- Automated, continuous backups with point‑in‑time recovery
- Choose highly‑available replicated pairs for business continuity
- Integrated metrics, logs, security and user controls at your fingertips
Spend less with compression savings from best‑in‑class algorithms, including delta-delta and Gorilla, and a memory‑efficient architecture.
- Reduce storage costs with 94-97% lossless compression rates
- Downsampling keeps aggregated metrics without wasting disk space
- Optimize storage consumption with data retention policies
- Transparent pricing, always know what you'll pay
- Dynamically scale compute and storage based on changing needs
This section describes the design and motivation around the TimescaleDB architecture, including why time-series data is different, and how we leverage its characteristics with TimescaleDB.
If you prefer to learn by watching and want an intro to TimescaleDB, check out our YouTube channel.
Found an issue on this page?Report an issue!