Hi, To get more accurate sampling, you need to bucketize your table based on the columns you wish to use in sampling. Also use the TABLESAMPLE clause while getting the required sample size in your queries.
http://hive.apache.org/docs/r0.9.0/language_manual/working_with_bucketed_tables.html https://cwiki.apache.org/Hive/languagemanual-sampling.html Regards Bejoy KS Sent from handheld, please excuse typos. -----Original Message----- From: Roberto Sanabria <robe...@stumbleupon.com> Date: Tue, 14 Aug 2012 15:31:14 To: <user@hive.apache.org> Reply-To: user@hive.apache.org Subject: Re: how to do random sampling in hive? Try this: select * from table_name order by rand() limit 5; Cheers, R On Tue, Aug 14, 2012 at 3:23 PM, Raihan Jamal <jamalrai...@gmail.com> wrote: > I think you can use here LIMIT- > > Limit indicates the number of rows to be returned. The rows returned are > chosen at random. The following query returns 5 rows from t1 at random. > > > > SELECT * FROM t1 LIMIT 5 > > http://karmasphere.com/hive-queries-on-table-data > > > > *Raihan Jamal* > > > > On Tue, Aug 14, 2012 at 3:18 PM, zuohua zhang <zuo...@gmail.com> wrote: > >> Would like to extract a uniform random sample from a hive table? How >> should I write the query? >> Thanks! >> > >
