Apache Kylin (Kylin) is a big data analysis engine contributed by eBay to the open source community. It supports second-level SQL and OLAP queries on large data sets. It is currently an incubation project of the Apache Foundation. Kylin uses Hadoop combined with data cube (Cube) technology to achieve multi-dimensional fast OLAP analysis capabilities.
- Scalable ultra-fast OLAP engine:
Kylin is designed to reduce query latency of tens of billions of data in Hadoop
- Hadoop ANSI SQL interface:
Kylin provides standard SQL for Hadoop to support most query functions
- Interactive query capability:
With Kylin, users can interact with Hadoop data in sub-seconds, providing better performance than Hive on the same dataset
- MOLAP Cube:
Users can define data models and build cubes in Kylin for more than 10 billion datasets
- Seamless integration with BI tools:
Kylin provides integration capabilities with BI tools such as Tableau, and will soon provide integration with other tools
- Other features:
- Job management and monitoring
- Compression and encoding
- Incremental update
--Use HBase Coprocessor
- Dinstinc Count approximation algorithm based on HyperLogLog
- Friendly web interface to manage, monitor and use the cube
- Item and cube level access control security
- Support LDAP
Prerequisites on Hadoop
Hadoop: 2.4+
Hive: 0.13+
HBase: 0.98+, 1.x
JDK: 1.7+
