Here is a possible solution

To add a root directory structure such as the following

as InputPath, do the following
For the output to mirror the input and to build based on Harsh J's response

You might be able to not use a reducer and use the MultipleOutput from the
mapper directly.

--
View this message in context: http://hadoop.6.n7.nabble.com/Running-hadoop-on-directory-structure-tp67904p68110.html
Sent from the common-user mailing list archive at Nabble.com.