[DRILL-4256] Performance regression in hive planning

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 1.5.0
• Fix Version/s: 1.5.0
• Component/s: Functions - Hive, Query Planning & Optimization
• Assignee: Venki Korukanti
• Reporter: Rahul Kumar Challapalli
• Reviewer: dgu-atmapr
• Created: 08/Jan/16 19:56
• Updated: 18/Feb/16 17:29
• Resolved: 20/Jan/16 18:54

Description

Commit # : 76f41e18207e3e3e987fef56ee7f1695dd6ddd7a

The fix for reading hive tables backed by hbase caused a performance regression. The data set used in the below test has ~3700 partitions and the filter in the query would ensure only 1 partition get selected.

Commit : 76f41e18207e3e3e987fef56ee7f1695dd6ddd7a
Query : explain plan for select count(*) from lineitem_partitioned where `year`=2015 and `month`=1 and `day` =1;
Time : ~25 seconds

Commit : 1ea3d6c3f144614caf460648c1c27c6d0f5b06b8
Query : explain plan for select count(*) from lineitem_partitioned where `year`=2015 and `month`=1 and `day` =1;
Time : ~6.5 seconds

Since the data is large, I couldn't attach it here. Reach out to me if you need additional information.