Hi, yiwang. If multiple cubes contain the same table, Kylin will do the following thing.
first, Kylin will choose the cube that contains all the dimensions in your SQL. second, if there are still more than one cubes, Kylin will choose the cube that has less cost(it depends on the dimension, measure and inner join number). 发送自 Windows 10 版邮件<https://go.microsoft.com/fwlink/?LinkId=550986>应用 ________________________________ 发件人: yiwang <756470...@qq.com> 发送时间: Friday, December 21, 2018 2:29:16 PM 收件人: dev@kylin.apache.org 主题: which cube is response for query when 2 models has different filters in 1 datasource table? Hello kylin team; for example, I created 2 Models using sample table: "KYLIN_SALES". Model: Test1 has filter seller_id=10000994, Test2 has filter seller_id=10000998 then I built 2 cubes for each model, get a valid table for query: KYLIN_SALES when i execute sql: select SELLER_ID,count(*) from KYLIN_SALES group by SELLER_ID all records returned are seller_id=10000994 which makes confused is one table contains multiple models and cubes, how is kylin to determine which model or cube should response for query? especially the same filter has different values like above what i said. Thanks Yi -- Sent from: http://apache-kylin.74782.x6.nabble.com/