What this book covers

Chapter 1, Introduction, gives the background to Hadoop and the Big Data problems it looks to solve. We also highlight the areas in which Hadoop 1 had room for improvement.

Chapter 2, Storage, delves into the Hadoop Distributed File System, where most data processed by Hadoop is stored. We examine the particular characteristics of HDFS, show how to use it, and discuss how it has improved in Hadoop 2. We also introduce ZooKeeper, another storage system within Hadoop, upon which many of its high-availability features rely.

Chapter 3, Processing – MapReduce and Beyond, first discusses the traditional Hadoop processing model and how it is used. We then discuss how Hadoop 2 has generalized the platform to use multiple computational models, of which MapReduce is merely one.

Chapter 4, Real-time Computation with Samza, takes a deeper look at one of these alternative processing models enabled by Hadoop 2. In particular, we look at how to process real-time streaming data with Apache Samza.

Chapter 5, Iterative Computation with Spark, delves into a very different alternative processing model. In this chapter, we look at how Apache Spark provides the means to do iterative processing.

Chapter 6, Data Analysis with Pig, demonstrates how Apache Pig makes the traditional computational model of MapReduce easier to use by providing a language to describe data flows.

Chapter 7, Hadoop and SQL, looks at how the familiar SQL language has been implemented atop data stored in Hadoop. Through the use of Apache Hive and describing alternatives such as Cloudera Impala, we show how Big Data processing can be made possible using existing skills and tools.

Chapter 8, Data Lifecycle Management, takes a look at the bigger picture of just how to manage all that data that is to be processed in Hadoop. Using Apache Oozie, we show how to build up workflows to ingest, process, and manage data.

Chapter 9, Making Development Easier, focuses on a selection of tools aimed at helping a developer get results quickly. Through the use of Hadoop streaming, Apache Crunch and Kite, we show how the use of the right tool can speed up the development loop or provide new APIs with richer semantics and less boilerplate.

Chapter 10, Running a Hadoop Cluster, takes a look at the operational side of Hadoop. By focusing on the areas of interest to developers, such as cluster management, monitoring, and security, this chapter should help you to work better with your operations staff.

Chapter 11, Where to Go Next, takes you on a whirlwind tour through a number of other projects and tools that we feel are useful, but could not cover in detail in the book due to space constraints. We also give some pointers on where to find additional sources of information and how to engage with the various open source communities.