MapReduce easily explained

Confused about the MapReduce algorithm for parallel processing of "Big Data"? It's an easy concept, e.g. used in Hadoop: http://java.dzone.com/articles/confused-about-mapreduce