[SPARK-13304] Broadcast join with two ints could be very slow

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Fix Version/s: 2.0.0
• Component/s: SQL
• Epic Link: Whole stage codegen
• Assignee: Davies Liu
• Reporter: Davies Liu
• Created: 12/Feb/16 20:46
• Updated: 20/Feb/16 06:58
• Resolved: 20/Feb/16 06:58

Description

If the two join columns have the same value, the hash code of them will be (a ^ b), which is 0, then the HashMap will be very very slow.

Issue Links

• links to: [Github] Pull Request #11188 (davies)