Chapter 5. Storage Component – HBase

One of the most important components of the Hadoop ecosystem is HBase, which utilizes HDFS very efficiently and can store, manage, and process data at a much better performing scale. NoSQL is emerging, and there is a lot of attention towards different implementations and solutions in Big Data problem solving space. HBase is a NoSQL database which can process the data over and above HDFS to achieve very good performance with optimization, scalability, and manageability. In Hadoop, HDFS is very good as storage for the WORM (Write Once Read Many) paradigm where data is not updated. In many scenarios, the requirements would be updating, ad hoc analysis or random reads. In HDFS, processing these requirements is not very efficient as updating a record in a file is not possible; HDFS has to delete and rewrite the whole file which is resource, memory and I/O intensive. But HBase can manage such processing efficiently in a huge volume of random read and writes...