Goun Na created HIVE-19283: ------------------------------ Summary: Select count(distinct()) a couple of times stuck in last reducer Key: HIVE-19283 URL: https://issues.apache.org/jira/browse/HIVE-19283 Project: Hive Issue Type: Improvement Components: CBO, Logical Optimizer Affects Versions: 2.1.1 Reporter: Goun Na Assignee: Ashutosh Chauhan
Distinct count query performance is significantly improved due to HIVE-10568. {code:java} select count(distinct elevenst_id) from 11st.log_table where part_dt between '20180101' and '20180131'{code} However, some queries that contain several distinct counts are still slow. It starts with multiple mappers, but stuck in the last one reducer. {code:java} select count(distinct elevenst_id) , count(distinct member_id) , count(distinct user_id) , count(distinct action_id) , count(distinct other_id) from 11st.log_table where part_dt between '20180101' and '20180131'{code} -- This message was sent by Atlassian JIRA (v7.6.3#76005)