Apache Kylin Interpreter for Apache Zeppelin
Overview
Apache Kylin is an open source Distributed Analytics Engine designed to provide SQL interface and multi-dimensional analysis (OLAP) on Hadoop supporting extremely large datasets, original contributed from eBay Inc. The interpreter assumes that Apache Kylin has been installed and you can connect to Apache Kylin from the machine Apache Zeppelin is installed.
To get start with Apache Kylin, please see Apache Kylin Quickstart.
Configuration
| Name | Default | Description |
|---|---|---|
| kylin.api.url | http://localhost:7070/kylin/api/query | kylin query POST API The format can be like http://<host>:<port>/kylin/api/query |
| kylin.api.user | ADMIN | kylin user |
| kylin.api.password | KYLIN | kylin password |
| kylin.query.project | learn_kylin | String, Project to perform query. Could update at notebook level |
| kylin.query.ispartial | true | true|false (@Deprecated since Apache Kylin V1.5) Whether accept a partial result or not, default be “false”. Set to “false” for production use. |
| kylin.query.limit | 5000 | int, Query limit If limit is set in sql, perPage will be ignored. |
| kylin.query.offset | 0 | int, Query offset If offset is set in sql, curIndex will be ignored. |
Using the Apache Kylin Interpreter
In a paragraph, use %kylin(project_name) to select the kylin interpreter, project name and then input sql. If no project name defined, will use the default project name from the above configuration.
%kylin(learn_project)
select count(*) from kylin_sales group by part_dt