Hadoop has a family of projects that are either built on top of Hadoop or work very closely with Hadoop. These projects have given rise to an ecosystem that focuses on large-scale data processing, and often, users can use several of these projects in combination to solve their use cases. This chapter introduces Apache Hive, which provides data warehouse capabilities on top of the data stored in HDFS. Chapter 7, Hadoop Ecosystem II – Pig, HBase, Mahout, and Sqoop introduces a few other key projects in the Hadoop ecosystem.

Apache Hive provides an alternative high-level language layer to perform large-scale data analysis using Hadoop. Hive allows users to map the data stored in HDFS into tabular models and process them using HiveQL, the SQL-like language layer, to query very large datasets using Hadoop. HiveQL can be used to perform ad-hoc querying of datasets as well as for data summarizations and to perform data analytics. Due to its SQL-like language, Hive is a natural choice...