[ZEPPELIN-3308] Accessing Hive table via JDBC Interpreter - Slow

• Type: Test
• Status: Open
• Priority: Critical
• Resolution: Unresolved
• Affects Version/s: 0.7.0
• Fix Version/s: 0.7.0
• Component/s: zeppelin-interpreter
• Labels: HIVE HQL JDBC zeppelin-server

Environment

Test environment

• Flags: Important
• Assignee: Unassigned
• Reporter: Mothilal
• Created: 07/Mar/18 20:29
• Updated: 09/May/18 10:38

Description

Simple Hive query : Select * from <default.sampleTable> limit 25 yielded response for 

Hive Client : <1 sec

Zeppelin %JDBC(Hive) : 30 sec.

While analyzing the logs, I understand it was taking time to fetch metadata of schema and did the same even when configuration specified default schema. Ideally the performance should be close to hive client with little latency but not major. Please advise if current data fetch approach is optimized for hive jdbc queries.