QuestDB
UK-based open-source time-series database built for high-throughput ingestion of timestamped data with SQL interface.
📍 United Kingdom 🇬🇧, London
Product overview
QuestDB is a time-series database built from scratch in Java, C++, and Rust with zero external dependencies. Designed for high-throughput ingestion of timestamped data, it uses columnar storage with SIMD-vectorized execution and time-partitioned architecture to deliver sub-second queries over billions of rows. The database supports SQL with time-series extensions and is compatible with PostgreSQL wire protocol and InfluxDB Line Protocol. Founded in 2019, QuestDB graduated from Y Combinator's Summer 2020 batch and raised approximately $14 million, including a $12 million Series A led by 468 Capital. Notable angel investors include the co-founders of GitHub, Docker, Reddit, and NGINX. The company operates with a lean remote-first team of roughly 25 engineers. The open-source edition, licensed under Apache 2.0, has earned over 16,400 GitHub stars. QuestDB Enterprise adds high availability, role-based access control, cold storage tiering to Apache Parquet on object storage, and advanced security features. QuestDB Cloud provides a fully managed deployment. Enterprise clients include Airbus (billions of data points daily), B3 (Brazilian stock exchange), and Aquis Exchange. Revenue is estimated at roughly $3 to $4 million. KEY FEATURES: - Purpose-built time-series engine with SIMD-vectorized columnar execution and zero-dependency architecture - Apache 2.0 open-source edition with 16,400+ GitHub stars and 50+ contributors - SQL interface with PostgreSQL wire protocol and InfluxDB Line Protocol for high-throughput ingestion - Multi-tier storage with automatic Parquet-based cold storage tiering to object stores - Deployed by Airbus, B3 (Brazilian stock exchange), and Aquis Exchange for billions of daily data points