Have you tried using external tables? BTW, hive tables can be defined as text tables, so you can run mapreduce on them too. Just locate the tables under directory: /user/hive/warehouse/<tablename>
Cheers, -Ajo. 2011/1/26 母延年YNM <muyann...@ku6.com> > When I use load data into table like this , > > LOAD DATA INPATH '/user/myname/kv2.txt' OVERWRITE INTO TABLE invites > PARTITION (ds='2008-08-15'); > > The raw data /user/myname/kv2.txt was rename into hive`s store direcrory > > > > > > Is there any method can not *move array the raw data*. > > > > Why not make a soft link? So that I can use both of the hive and > hadoop-mapreduce. >
