In this short course offered by Udacity, I learnt the fundamentals of MapReduce and Apache Hadoop to start making sense of Big Data in the real world!
Every assignment folder has a 'testfile' that acts as a local subset of the main dataset on the server. The Mapper and Reducer codes are first tested on this file and then run on the main dataset.
The comment section of each mapper code talks about its particular problem statement.