This is a video  I found from youtube that explains hadoop mapreduce very clearly using the wordcount example.