Apache Hadoop Yarn - 9780321934505

This authoritative guide is the best source of information for getting started and mastering the latest advancements in Apache Hadoop. As you learn how to structure your applications in Apache Hadoop 2, it provides an understanding of the architecture of YARN (code name for Hadoop 2) and its major components. Including multiple examples and valuable case studies, a key topic is running existing Hadoop 1 applications on YARN and the MapReduce 2 infrastructure.

Data processing in Apache Hadoop has undergone a complete overhaul to emerge as Apache Hadoop YARN. This generic compute fabric provides resource management at datacenter scale, and a simple method to implement distributed applications (e.g. MapReduce and a multitude of others) to process petabytes of data on Apache Hadoop HDFS. YARN significantly changes the game to recast Apache Hadoop as a much more powerful system by moving it beyond MapReduce. Through this work, some of the primary authors of the YARN project such as Arun C. Murthy, the Founder of the YARN project, and Vinod K. Vavilapalli, the YARN Project Lead, take the reader through the key design concepts for YARN itself and also on a tour of how new applications can be written in an elegantly simple manner to get more out of Hadoop clusters as Hadoop is no longer a one-trick pony. Learn how existing MapReduce applications can be seamlessly migrated to YARN in a hassle-free manner, and how other existing components in Apache Hadoop ecosystem such as Apache Hive, Apache Pig & Apache HBase improve thanks to YARN.