# apache/kyuubi

Apache Kyuubi is a distributed and multi-tenant gateway to provide serverless SQL on data warehouses and lakehouses.

Repository: https://github.com/apache/kyuubi
Canonical: https://ross.abutalabs.com/products/kyuubi
Homepage: https://kyuubi.apache.org/
Language: Scala
License: Apache-2.0
License Family: permissive
Topics: spark, hive, sql, thrift, jdbc, spark-sql, data-lake, hadoop, kubernetes, hacktoberfest
Last push: 2026-08-26T20:56:15+00:00

## Health v2 (maintenance only)
Score: 94/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 83, longevity 100
- inputs: {"age_days": 3180, "days_push": 7, "days_rel": 36, "gap_med": 40.0, "n_releases_24m": 11}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2361, forks 1017 (observed 2026-08-28T04:06:40.907293+00:00)

## What it is
Apache Kyuubi is a distributed, multi-tenant Thrift JDBC/ODBC gateway that provides serverless SQL access to data warehouses and lakehouses. It manages pre-warmed Spark, Flink, and Trino engine instances so end users and BI tools can run SQL against large-scale data without managing compute themselves.

## Use cases
- run interactive SQL queries on Spark from BI tools via JDBC
- provide a multi-tenant SQL gateway for a data lakehouse
- serve HiveServer2-compatible connections for big data analytics
- run large ETL batch jobs through a unified SQL interface
- isolate compute resources per user or tenant on a shared cluster
- query Iceberg, Hudi, Delta, and Hive tables from one endpoint

## When to choose
- you need a serverless, multi-tenant SQL entry point over Spark/Flink/Trino
- you want HiveServer2 compatibility so existing JDBC/ODBC clients and BI tools work unchanged
- you require per-user engine isolation, high availability, and high client concurrency
- you want unified authentication and authorization across multiple query engines

## When to avoid
- you need a lightweight embedded SQL engine rather than a distributed gateway service
- your workloads are small-scale single-user queries where a direct Spark or Trino connection suffices
- you need OLTP-style transactional queries against a relational database
- you cannot operate JVM-based server infrastructure or a Hadoop/lakehouse storage layer

## Facets
- artifact type: service
- maturity: stable
- function: database, api-framework, auth, authorization, scheduling
- domain: big-data, databases, analytics, microservices
- platform: jvm, cloud, self-hosted
- tags: spark-sql, jdbc-odbc-server, lakehouse, hive, serverless-sql, multi-tenancy, etl-gateway, apache-software-foundation, sql, data-engineering, kubernetes, docker, linux

## Member repositories
- apache/kyuubi (main) score 94

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:40.907293+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:36:06.151784+00:00, confidence not recorded.
  - readme: https://github.com/apache/kyuubi (fetched 2026-08-28T04:06:40.907293+00:00, sha 0a435d50f430)
  - homepage: https://kyuubi.apache.org/ (fetched 2026-08-29T10:16:38.486761+00:00, sha 0ba0cb360f74)
  - site_page: https://kyuubi.apache.org/releases.html (fetched 2026-08-29T10:16:38.495909+00:00, sha 23bc75780c74)
- Data as of 2026-08-30T08:39:29.467469+00:00.