[HIVE-12369] Native Vector GROUP BY

• Type: Improvement
• Status: Patch Available
• Priority: Critical
• Resolution: Unresolved
• Component/s: Hive
• Assignee: Matt McCline
• Reporter: Matt McCline
• Created: 09/Nov/15 10:25
• Updated: 21/Dec/23 20:13

Description

Implement Native Vector GroupBy using fast hash table technology developed for Native Vector MapJoin, etc.

Patch is currently limited to a single COUNT aggregation.  Or, no aggregation also known as duplicate reduction.

Here are examples of new kinds of classes introduces that stored the count in the slot table and don't allocate hash elements:

  COUNT(column)  VectorGroupByHashLongKeySingleCountColumnOperator      
  COUNT(key)     VectorGroupByHashLongKeySingleCountKeyOperator            
  COUNT(*)       VectorGroupByHashLongKeySingleCountStarOperator           

And the duplicate reduction operator a single key.  Example:

  VectorGroupByHashLongKeyDuplicateReductionOperator

Issue Links

• links to: RB #68898