Chapter 9

Distributed and Parallel Computing

The Scala Actors API puts a powerful, yet simple, parallel computing framework built on top of the JVM at your disposal. This chapter illustrates how to accomplish some common parallel and distributed computing tasks with actors. In particular, we focus on two patterns that are useful in many applications: MapReduce and reliable broadcasting. MapReduce is a paradigm for parallel and distributed programming that has been established as a de facto standard to accomplish a wide variety of tasks, such as hypertext document processing, machine learning, and data mining. Reliable broadcasting, on the other hand, is often necessary in distributed applications where machines in a cluster can fail due to hardware outages or communication delays.