[ https://issues.apache.org/jira/browse/HIVE-10334?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14495150#comment-14495150 ]
Sergey Shelukhin commented on HIVE-10334: ----------------------------------------- That is because column 0 (the only one read) has no data streams > LLAP: Self-joins with the cache enabled > --------------------------------------- > > Key: HIVE-10334 > URL: https://issues.apache.org/jira/browse/HIVE-10334 > Project: Hive > Issue Type: Sub-task > Affects Versions: llap > Reporter: Gopal V > Assignee: Sergey Shelukhin > > Looks like reading the same split from 2 tasks in the same process might have > some corner cases. > {code} > hive> select count(1) from tpch_orc_snappy_1000.region y join > tpch_orc_snappy_1000.region x; > OK > 0 > {code} > {code} > hive> set hive.llap.io.enabled=false; > hive> select count(1) from tpch_orc_snappy_1000.region y join > tpch_orc_snappy_1000.region x; > OK > 25 > {code} -- This message was sent by Atlassian JIRA (v6.3.4#6332)