Hi all, I am new to Hadoop Development and am working on a project which involves writing the intermediate Map output to a Parallel File system(Lustre) and tweaking the Reducer to read from the same during shuffle phase.
My doubt is : *What classes do I need to look for to solve the above stated problem?* I went through couple of books but couldn't find much detailed information. Looking into the source code I felt it must be the OutputCollector Class. Please correct me if am wrong. Any help or pointers are highly appreciated.Thanks. -- --With Regards Pavan Kulkarni
