MapReduce

Hadoop's parallel processing engine that can process large data sets relatively quickly. A necessary component in addition to MapReduce in a Hadoop distribution.