In this chapter, we covered the architecture and implementation of HDFS Federation, which is useful for splitting the namespace into multiple parts and managing them separately. Then we implemented HDFS high availability by configuring two namenodes as an active/passive configuration. We also configured jobtracker high availability by configuring two jobtrackers in an active/passive configuration. Both HDFS HA and jobtracker HA can be configured for manual failover or automatic failover using Apache ZooKeeper.

In the next chapter, we will learn the architecture and implementation of Cloudera Manager, Cloudera's Apache Hadoop administration tool. We will cover all of its features and how it can be used to administer a cluster running CDH5.