The MapReduce framework promises to make computing of large sets of data very easy. An original approach, it offers excellent scalability across many computing nodes, and can easily be integrated with existing systems. This session will give an introduction to the basic techniques and ideas behind MapReduce, followed by hands-on examples using Apache Hadoop, a major implementation of MapReduce, including Hadoop's streaming functionality that allows users to write mapper and reducer tasks in PHP rather than Java.

Comments

Comments are closed.