Hadoop cluster

A Hadoop cluster is a specialized cluster for storing and analyzing enormous amounts of unstructured data in a distributed computing environment.

Corresponding clusters work with Hadoop's open source distributed systemssoftware running on inexpensive computers. The concept works in master- slave mode, where one computer forms the NameNode, and another forms the JobTracker. These two nodes are master nodes. The user data is stored in the slaves, which operate as DataNodes or TaskTrackers.

Hadoop clusters increase the processing speed for data analysis, they are scalable, their computational and storage power can be increased at any time by adding more cluster nodes. They have high fault redundancy because all data is stored on multiple cluster nodes.