set hive.auto.convert.join = true; -- EXCLUDE_HADOOP_MAJOR_VERSIONS( 0.20S) CREATE TABLE dest1(c1 INT, c2 STRING) STORED AS TEXTFILE; set mapreduce.framework.name=yarn; set mapreduce.jobtracker.address=localhost:58; set hive.exec.mode.local.auto=true; explain FROM src JOIN srcpart ON src.key = srcpart.key AND srcpart.ds = '2008-04-08' and src.key > 100 INSERT OVERWRITE TABLE dest1 SELECT src.key, srcpart.value; FROM src JOIN srcpart ON src.key = srcpart.key AND srcpart.ds = '2008-04-08' and src.key > 100 INSERT OVERWRITE TABLE dest1 SELECT src.key, srcpart.value; SELECT sum(hash(dest1.c1,dest1.c2)) FROM dest1;