Apache Kylin 1.3.0 发布， OLAP 分析引擎
欢迎加入运维网交流群：263444886Apache Kylin 1.3.0 发布了，Kylin 是一个开源的分布式的 OLAP 分析引擎，来自 eBay 公司开发，基于 Hadoop 提供 SQL 接口和 OLAP 接口，支持 TB 到 PB 级别的数据量。
Kylin 是：
[*] 超级快的 OLAP 引擎，具备可伸缩性
[*] 为 Hadoop 提供 ANSI-SQL 接口
[*] 交互式查询能力
[*] MOLAP Cube
[*] 可与其他 BI 工具无缝集成，如 Tableau，而 Microstrategy 和 Excel 将很快推出
其他值得关注的特性包括：
[*] 作业管理和监控
[*] 压缩和编码的支持
[*] Cube 的增量更新
[*] Leverage HBase Coprocessor for query latency
[*] Approximate Query Capability for distinct Count (HyperLogLog)
[*] 易用的 Web 管理、构建、监控和查询 Cube 的接口
[*] Security capability to set ACL at Cube/Project Level
[*] 支持 LDAP 集成
改进日志：
新特性
* - Unload table from Kylin
* - Support Custom Aggregation Types
* - Support Hive client Beeline
* - Clone Cube Metadata
* - Support precise Count Distinct using bitmap (under limited conditions)
提升
* - HiveColumnCardinalityJob should use configurations in conf/kylin_job_conf.xml
* - Support kerberos authentication while getting status from RM
* - Load hive tables with selecting mode
* - Hive dependencies should be add to tmpjars
* - make filtering input easier in creating cube
* - Enhance project level ACL
* - Add 'Enable Cache' button in System page
* - Cube ACL does not work
* - Fix link and typo in README
* - In query window, enable fast copy&paste by double clicking tables/columns' names.
* - get cubes API update - offset,limit not required
* - add toggle event for tree label
* - Change font/background color of job progress
* - Enhance DeployCoprocessorCLI to support Cube level filter
* - Kill underlying running hadoop job while discard a job
* - Improve performance of converting data to hfile
* - Kylin Entity Permission Control
* - Upgrade calcite version to 1.6
* - Kylin ACL enhancement
* - JDBC Driver is not generic to restAPI json result
Bug 修复
* - Calcite throws "java.lang.Float cannot be cast to java.lang.Double" error while executing SQL
* - select from is not supported
* - Cannot have comments in the end of New Query textbox
* - Long dimension value cause ArrayIndexOutOfBoundsException
* - can not see project options after clear brower cookie and cache
* - problem about kylin web UI
* - Remove "Back to My Cubes" link in non-edit mode
* - minor, update website member's info on community page
* - When CubeMigrationCLI copied ACL from one env to another, it may not work
* - redirect to home page when input invalid url
* - Got NPE when discarding a job
* - cube model will be overridden while creating a new cube with the same name
* - Job status labels are not in same style
* - Query from JDBC is partial results by default
* - Wrong label in Dialog CUBE REFRESH CONFIRM
* - UI create model: Press enter will go back to pre step
* - UI Delete Aggregation Groups: cursor disappeared after delete 1 dimension
* - Typo in doc
* - Couldn't edit a cube if it has no "partition date" set
* - Query using PrepareStatement failed with multi OR clause
* - minor bug in BigDecimalSerializer - avoidVerbose should be incremented each time when input scale is larger than given scale
* - kylin.metadata.url with hbase namespace problem
* - StringIndexOutOfBoundsException in Kylin Hive Column Cardinality Job
* - Widget> * - NullPointerException occurs when query from subqueries with order by
* - HBase>
* - For setting Auto Merge Time Ranges, before sending them to backend, the> * - Kylin should throw error if HIVE_CONF dir cannot be found
* - Shouldn't use "1970-01-01" as the default end date
* - Some environment variables are not used in bin/kylin.sh
* - Hive dependency jars are hard coded in test
详情请看：https://kylin.apache.org/docs/release_notes.html
下载地址：https://kylin.apache.org/download/
页:
[1]