Fwd: [ANNOUNCE] Apache Druid 37.0.0 release

[YQ Mei]

---------- Forwarded message ---------
From: Cece Mei <y...@apache.org>
Date: Fri, May 8, 2026 at 3:39 PM
Subject: [ANNOUNCE] Apache Druid 37.0.0 release
To: <d...@druid.apache.org>, <druid...@googlegroups.com>

The Apache Druid team is proud to announce the release of Apache Druid 37.0.0. Druid is a high performance analytics data store for event-driven data.

Apache Druid 37.0.0 contains over 255 new features, performance enhancements, bug fixes, and major documentation improvements from 29 contributors. Major new features and improvements include:

- Removed hadoop-based ingestion
- Dynamic query blocklist
- Dynamic default query context on Broker
- Broker tier selection for real-time servers
- Manual Broker routing in the web console
- Minor compaction to compact only fragmented segments for Overlord-based compaction
- Fingerprinting mechanism to track compaction states for Overlord-based compaction
- Cascading reindexing to apply different compaction configurations based on the age of your data
- Read-only authorizer
- Thrift input format
- sys.queries table for Dart engine
- New Consul contrib extension for service discovery and Coordinator/Overlord leader election

Also, Auto-compaction using compaction supervisors, Multi-supervisor ingestion, Incremental cache, Kubernetes-based task management are now generally available!

Source and binary distributions can be downloaded from:
https://druid.apache.org/downloads.html

Release notes are at:
https://github.com/apache/druid/releases/tag/druid-37.0.0

A big thank you to all the contributors in this milestone release!