Use Bucketing using ID and use sort , also store as ORC file format and compress and set hive.exec.orc.split.strategy=BI; Example : CREATE TABLE Employee( ID BIGINT, NAME STRING, AGE INT, SALARY BIGINT, DEPARTMENT STRING ) COMMENT 'This is Employee table clustered by id sorted by age into 5 buckets' CLUSTERED BY(ID) SORTED BY(AGE)INTO 5 BUCKETS STORED AS ORC tblproperties (“orc.compress” = “SNAPPY”); Let me know if you face any issue ... View more