[ https://issues.apache.org/jira/browse/HIVE-18866?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16478452#comment-16478452 ]
Sergey Shelukhin commented on HIVE-18866: ----------------------------------------- Added a test and fixed the short impl that was wrong for some reason (even after fixing 32 to 16). > Semijoin: Implement a Long -> Hash64 vector fast-path > ----------------------------------------------------- > > Key: HIVE-18866 > URL: https://issues.apache.org/jira/browse/HIVE-18866 > Project: Hive > Issue Type: Improvement > Components: Vectorization > Reporter: Gopal V > Assignee: Sergey Shelukhin > Priority: Major > Labels: performance > Attachments: 0001-hash64-WIP.patch, HIVE-18866.patch, > perf-hash64-long.png > > > A significant amount of CPU is wasted with JMM restrictions on byte[] arrays. > To transform from one Long -> another Long, this goes into a byte[] array, > which shows up as a hotspot. > !perf-hash64-long.png! -- This message was sent by Atlassian JIRA (v7.6.3#76005)