java - Hadoop: intervals and JOIN -

I am very new to Hadop and I am currently in the data of two sources Trying where the key is a interval (say [date-start / date-end]). For example:

Input 1:

  20091001-20091002 A 20091011-20091104b 2008-0111-20091103C (...)  

Input 2:

  20090902-20091003D 20081015-20091204E 20040011-20050101F (...)  

I should get all the records Key1 Key 2 overlaps Is it possible? Where can I find an example of implementation?

Thank you.