Apache Druid is a free, open-source database built for sub-second analytics queries on streaming and batch data at massive scale.
Apache Druid is an open-source, real-time analytics database purpose-built for sub-second queries over streaming and batch data at scale. It was originally developed in 2011 by Eric Tschetter, Fangjin Yang, Gian Merlino, and Vadim Ogievetsky at the advertising analytics startup Metamarkets, released as open source in 2012, and became an Apache Software Foundation project in 2015 under the Apache License 2.0.
As an Apache project, Druid is maintained by an open-source community rather than a single vendor, though commercial companies such as Imply provide managed hosting and support built on top of it.
Druid stores data in a columnar, time-indexed, and compressed format optimized for analytical queries, and natively ingests streaming data from sources such as Apache Kafka and Amazon Kinesis alongside batch data. It supports standard SQL for querying in addition to its native JSON-based query language, and is designed to sustain high query concurrency, from hundreds to hundreds of thousands of queries per second, with sub-second latency.
Druid's architecture separates ingestion, storage, and query-serving into loosely coupled components, which allows clusters to scale elastically as data volume and query load grow. Built-in data replication across nodes provides fault tolerance and continuous availability for production workloads.
Yes, Druid is fully open source under the Apache License 2.0 and free to self-host. Commercial managed offerings are available from third-party vendors such as Imply.
Druid is used to power real-time analytics dashboards, clickstream and event analytics, and operational monitoring systems that need sub-second queries over large volumes of data.
Yes, Druid supports standard SQL queries in addition to its native JSON-based query language.
Druid is maintained as an Apache Software Foundation project by an open-source community. It was originally created in 2011 at Metamarkets before being open-sourced and later donated to the Apache Software Foundation.
Yes, Druid natively integrates with streaming sources such as Apache Kafka and Amazon Kinesis for real-time ingestion.