What is the value of "COLUMN_STATS_ACCURATE" field in "DESCRIBE FORMATTED <table_name>" output under "Table Parameters" section. My suspect is, it's false even after executing your analyze query. If so, Hive launches the mapred jobs.
Try using below analyze query: analyze table mytable partition(server_date=’2013-11-30′) compute statistics; --Bala G. On Tue, Jul 22, 2014 at 10:14 AM, Navdeep Agrawal < [email protected]> wrote: > i am trying to compute statistics on ORC File but i am unable see any > changes in PART_COL_STATS as well on using > > set hive.compute.query.using.stats=true; > set hive.stats.reliable=true; > set hive.stats.fetch.column.stats=true; > set hive.stats.fetch.partition.stats=true; > set hive.cbo.enable=true; > > to get max value of a column it is running full Map reduce on column .. > what i want to use is max value stored in meta store ,but i am unable to > catch these statistics . > > my table desc is > load_inst_id int > src_filename string > server_date date > > my analyze query is > analyze table mytable partition(server_date=’2013-11-30′) compute > statistics for columns load_inst_id; > > i am always getting 0 as loadinstant id ,i have to turn off my > hive.compute.query.using.stats to get correct result(through map reduce > max(load_inst_id)) > > >
