Saturday, January 4, 2014

Big Data: What is a Rack in Hadoop Cluster?

A Node is a single computer in Hadoop cluster.

A Rack is a collection of nodes; usually 30 to 40 nodes. All the nodes in a Rack share the same network switch. All the nodes are physically located close to each other. The bandwidth speed among the nodes in a Rack is quick.

A node in a Rack can write or read data quickly from the node located the same rack. A rack when read or write data from a different rack is usually slow. The speed is not noticeable for technically it happens.

Please visit the Big Data page to read all articles on Big Data

Popular Posts

Real Time Web Analytics