100 millions is very high cardinality, we have dictionary threshold for this, please search in JIRA or this mailing list, There's discussion previous.
Thanks. Best Regards! --------------------- Luke Han On Thu, Aug 6, 2015 at 8:13 AM, liangmeng <[email protected]> wrote: > should i select false and fill the length of the dimension string length > at "Advanced Setting" when defining a cube? > what's the meaning for these two? > > ------------------------------ > 梁猛 > 中国移动广东公司 网管维护中心 网管支撑室 > 电话：13802880779 > 邮箱: [email protected] ，[email protected] > 地址：广东省广州市珠江新城珠江西路11号 广东全球通大厦北3楼 > 邮编：510623 > > > *发件人：* liangmeng <[email protected]> > *发送时间：* 2015-08-06 00:56 > *收件人：* dev <[email protected]> > *主题：* exception at step 3:Build Dimension Dictionary > i define a normal dimension from fact table which might have distinct > values up to 100 millions, but the cube build failed at step 3, error log > as below: > > > java.lang.IllegalArgumentException: Too high cardinality is not suitable for > dictionary -- DEFAULT.LTE_HTTP_ETL.MSISDN cardinality: 7908567 > > at > org.apache.kylin.dict.DictionaryGenerator.buildDictionaryFromValueList(DictionaryGenerator.java:90) > > at > org.apache.kylin.dict.DictionaryGenerator.buildDictionary(DictionaryGenerator.java:125) > > should i make a new lookup table to save these value or has some other > ways? > ------------------------------ > 梁猛 > 中国移动广东公司 网管维护中心 网管支撑室 > 电话：13802880779 > 邮箱: [email protected] ，[email protected] > 地址：广东省广州市珠江新城珠江西路11号 广东全球通大厦北3楼 > 邮编：510623 > >
