Repolode
Submit
StarRocks

StarRocks/starrocks

An open-source MPP query engine for sub-second analytics on data lakehouses and in its own storage, speaking the MySQL protocol.

Open-source alternative to:SnowflakeGoogle BigQuery

View on GitHubWebsite
Stars
12,241
Forks
2,655
Open issues
1,721
Last commit
23 hours ago

What is it?

StarRocks is a distributed SQL query engine built for analytics workloads, including multi-dimensional analysis, real-time analytics and ad-hoc queries. It runs in both modes: querying data in place in lakehouse formats such as Apache Iceberg, Delta Lake, Apache Hudi and Apache Hive without importing it, and serving data stored in its own tables.

The engine is written in Java and C++ and uses a vectorized execution model. Its architecture consists of Frontend (FE) and Backend (BE) nodes that scale horizontally and replicate metadata and data. From version 3.0 it also offers a shared-data architecture. Tables with primary-key models support upserts and deletes, and materialized views can be refreshed during data import and selected automatically at query time.

StarRocks supports ANSI SQL and the MySQL protocol, so existing MySQL-compatible clients and BI tools can connect to it. Resource management lets operators limit query resource consumption and isolate tenants within one cluster.

Key features

  • 01Vectorized execution engine with a cost-based optimizer
  • 02ANSI SQL support and MySQL protocol compatibility
  • 03Real-time upsert and delete on primary-key tables
  • 04Automatically maintained and selected materialized views
  • 05Direct querying of Iceberg, Delta Lake, Hudi and Hive data without import
  • 06Resource limits and isolation for queries and tenants

Use cases

  • Building real-time dashboards and customer-facing analytics over fresh data
  • Running ad-hoc multi-dimensional queries on a data lakehouse without moving data
  • Replacing a denormalized wide-table workflow with star-schema joins
  • Connecting existing MySQL-compatible BI tools to a larger analytical dataset

Quick start

  1. 1

    Download StarRocks from the community download page

  2. 2

    Follow the Quick Starts guide in the docs to deploy it

Things to know

  • Running it requires deploying Frontend (FE) and Backend (BE) components, so self-hosting takes setup effort.

In collections

Databases & SQL tools

Topics

#olap#analytics#sql#lakehouse#real-time-analytics#data-lake#distributed-database#mpp#big-data#cloudnative

FAQ

What is starrocks?
An open-source MPP query engine for sub-second analytics on data lakehouses and in its own storage, speaking the MySQL protocol. StarRocks is a distributed SQL query engine built for analytics workloads, including multi-dimensional analysis, real-time analytics and ad-hoc queries. It runs in both modes: querying data in place in lakehouse formats such as Apache Iceberg, Delta Lake, Apache Hudi and Apache Hive without importing it, and serving data stored in its own tables.
Is starrocks free to use? What license does it have?
starrocks is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
Is starrocks still maintained?
Actively maintained. The last commit to starrocks was on Nov 6, 2026.
What can I use starrocks for?
Building real-time dashboards and customer-facing analytics over fresh data Running ad-hoc multi-dimensional queries on a data lakehouse without moving data Replacing a denormalized wide-table workflow with star-schema joins Connecting existing MySQL-compatible BI tools to a larger analytical dataset
How do I install starrocks?
Download StarRocks from the community download page Follow the Quick Starts guide in the docs to deploy it
Where can I get starrocks?
The source code and installation instructions are on GitHub: https://github.com/StarRocks/starrocks

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects