hi,all. I'm using cdh5.13.1. i have two  parquet table: p_t_customer(3billion rows) and p_t_contract_product(4billion rows). i use this simple join sql:   select count(*) from p_t_customer c inner join p_t_contract_product p on c.customer_id = p.insured_2; but when i use hive on spark,it is not stop after hours,but when i use sparksql ,it executes succesfull in 15miutes.   here is hive on spark screenshot: ... View more