I created a ORC table A: *create table A (id string i, age int, num_children int);* *alter table A set fileformat orc;* *insert into table A select * from B cluster by age;*
But, When I run query, *select id from A where age=60;* The data read from HDFS is almost equal to the size of A. I expected lesser data due to predicate pushdown. No improvement in time also. Is there a setting Im missing. I run Hive 0.13 and Hadoop 2.7.0-mapr-1506. Thanks.