Open-source time-series database for low-latency SQL, high-throughput ingestion, and open formats
- Stars17.3k
- Forks1.6k
- Open Issues952
Apache-2.0
- Java
- Rust
- C++
About QuestDB
QuestDB is an open-source time-series database for workloads that need fast ingestion and low-latency SQL over time-ordered data. It is built for financial market data, sensor and telemetry data, and real-time dashboards and monitoring, with open formats and no vendor lock-in.
It uses a multi-tier storage engine with WAL, native columnar storage, and Parquet on object storage. The core engine is implemented in zero-GC Java and C++, with SIMD instructions, parallel vector execution, a column-oriented storage model, time-series SQL extensions, a Web Console, PGwire, REST API, and materialized views.
QuestDB ships with Docker and Homebrew install paths and supports first-party ingestion clients for Python, .NET, C/C++, Go, Java, NodeJS, and Rust. It is Apache 2.0 licensed, and QuestDB Enterprise adds high availability, read replicas, multi-primary ingestion, cold storage integration, role-based access control, TLS encryption, and support SLAs.
Key features
- Low-latency SQL with ASOF JOIN, WINDOW JOIN, HORIZON JOIN, SAMPLE BY, and LATEST ON
- High-throughput ingestion from single events to millions/sec
- Multi-tier storage: WAL, native columnar, and Parquet on object storage
- Web Console, PostgreSQL Wire Protocol, and REST API
- Materialized views, n-dimensional arrays, and finance functions
Details
- First released
- 2014
- Platforms
- Web · Docker · CLI
- Deployment
- self-hostable · docker · cloud
- Storage
- WAL · native · Parquet
- License
- Apache 2.0
- Language
- Java · C++ · Rust