Apache Spark is beyond the scope of this book, so if you want to know more about this powerful framework, I suggest you read the online documentation or one of the many books available. In Machine Learning with Spark, Pentreath N, Packt Publishing, there's an interesting introduction to the Machine Learning Library (MLlib) and how to implement most of the algorithms discussed in this book.
Spark is a parallel computational engine that is now part of the Hadoop project (even if it doesn't use its code), which can run in local mode or on very large clusters (with thousands of nodes) to execute complex tasks using huge amounts of data. It's mainly based on Scala, though there are interfaces for Java, Python, and R. ...
