whichdb

Apache Kylin

active

by Apache Software Foundation · since 2014 · also known as Kylin

OLAP engine that pre-computes cubes and aggregate indexes for sub-second SQL on huge datasets.

columnar / olaprelationaldatabase engineopen source (permissive)mature
Website Docs Source★ 3.8K Compare Find similar

Overview

Apache Kylin, open-sourced by eBay in 2014, answers SQL over billions of rows in sub-second time by pre-computing aggregates: you define a model (fact and dimension tables), Kylin builds cube or aggregate indexes with Spark, and queries are routed to the smallest matching index. Kylin 5 reworked this around flexible indexes, internal tables, a Spark-based query engine and Parquet storage. It suits stable, dimensional BI workloads with predictable queries; ad-hoc analysis and fast-changing schemas fit modern MPP engines better. Kyligence sells the commercial version.

Analytics (OLAP)Data warehouse

Strengths & trade-offs

Strengths

  • +Sub-second queries on huge datasets via pre-computed aggregate indexes
  • +Predictable latency for dashboards regardless of raw data size
  • +Standard SQL and BI connectors (Tableau, Power BI, Excel MDX via Kyligence)
  • +Proven at eBay, Meituan and other large internet companies

Trade-offs

  • −Requires upfront modeling and index builds; poor for ad-hoc queries
  • −Index build time and storage grow with dimensions
  • −Small, mostly China-based community and slow releases

Great for

  • +Enterprise BI dashboards with fixed dimensions
  • +Replacing legacy MOLAP cubes
  • +High-concurrency reporting

Think twice for

  • −Ad-hoc exploration
  • −Real-time streaming analytics
  • −Small datasets

Capabilities

Native or official first-party support · 6 of 25 tracked

✓Joins
✓Materialized views
✓Columnar storage
✓Durable persistence
✓Compression
✓Role-based access
–Transactions
–Secondary indexes
–Full-text search
–Vector search
–Geospatial
–JSON
–Change data capture
–Triggers
–Stored procedures
–TTL / expiry
–Pub/Sub
–In-memory
–Encryption at rest
–Time travel
–Branching
–Online schema changes
–Streaming ingest
–Offline sync
–Runs in WebAssembly

Scale & consistency

Comfortable data size
Massive · Multi-PB
Horizontal scaling
Native
Multi-region
No
Transactions
None
Consistency
Eventual
Schema
Strict
Architecture
Distributed, Client server
Query languages
SQL (ANSI)

Cube builds run on Spark; query nodes scale out; trillion-row deployments reported by users.

Deployment

Self-hosted Vendor-managed cloud Serverless Embedded Edge Kubernetes operator

Ecosystem

apis & protocolsREST API, JDBC, ODBC
client languagesJava, Python

Alternatives to Apache Kylin

Compare all→
Apache DruidColumnar / OLAP35.0

Real-time analytics database for sub-second slice-and-dice over streaming event data.

apache-kylin vs apache-druid →
StarRocksColumnar / OLAP15.6

MPP OLAP database with a fast vectorized engine, real-time upserts and native lakehouse queries.

apache-kylin vs starrocks →
ClickHouseColumnar / OLAP69.8

Open-source columnar OLAP database known for very fast aggregations over billions of rows.

apache-kylin vs clickhouse →
Apache DorisColumnar / OLAP15.9

Real-time MPP analytical database with MySQL compatibility, search indexes and lakehouse access.

apache-kylin vs apache-doris →
Apache PinotColumnar / OLAP29.9

Real-time OLAP store from LinkedIn for user-facing analytics at very high query rates.

apache-kylin vs apache-pinot →
SAP IQColumnar / OLAP11.9

Sybase's pioneering columnar warehouse, now SAP IQ and the engine behind HANA Cloud's data lake.

apache-kylin vs sap-iq →

Apache Kylin FAQ

What is Apache Kylin?

Apache Kylin is a columnar / OLAP database by Apache Software Foundation, first released in 2014. Apache Kylin, open-sourced by eBay in 2014, answers SQL over billions of rows in sub-second time by pre-computing aggregates: you define a model (fact and dimension tables), Kylin builds cube or aggregate indexes with Spark, and queries are routed to the smallest matching index. Kylin 5 reworked this around flexible indexes, internal tables, a Spark-based query engine and Parquet storage. It suits stable, dimensional BI workloads with predictable queries; ad-hoc analysis and fast-changing schemas fit modern MPP engines better. Kyligence sells the commercial version.

Is Apache Kylin open source?

Yes. Apache Kylin is open source. License: Apache License 2.0.

Does Apache Kylin support ACID transactions?

No. Apache Kylin does not provide ACID transactions; its consistency model is eventual.

Does Apache Kylin scale horizontally?

Apache Kylin scales horizontally out of the box, with built-in sharding or partitioning. It is comfortable at massive scale (Multi-PB).

How can I run Apache Kylin?

You can self-host it.

What is Apache Kylin best for?

Apache Kylin is a strong fit for enterprise BI dashboards with fixed dimensions, replacing legacy MOLAP cubes and high-concurrency reporting. Think twice for ad-hoc exploration.

What are the best alternatives to Apache Kylin?

Common alternatives to Apache Kylin include Apache Druid, StarRocks, ClickHouse, Apache Doris and Apache Pinot. The right choice depends on your workload, scale and hosting; WhichDB's smart match ranks them against your requirements.

How popular is Apache Kylin?

Apache Kylin ranks #196 of 844 databases on the WhichDB Score (16.9/100), and #19 among columnar / OLAP databases. The score combines public adoption signals such as package downloads, GitHub activity, job posts and developer discussion.