If you’ve been curious about Hadoop, the open source cloud computing software, there’s a really great set of introductory lectures being provided by Cloudera. In addition to building and supporting Hadoop clusters for commercial applications, the Cloudera folks have also been providing a wealth of information on installing and optimizing Hadoop on your own hardware, as well as a set of utilities for easily launching and managing a Hadoop cluster on EC2.
In the first part of the training series, embedded above, Aaron Kimball gives an introduction to large scale data processing, its challenges, and how a system like Hadoop addresses the issues commonly encountered when performing data computation across thousands of nodes. Other lectures include installing Hadoop, writing MapReduce programs, and using Hive, a data warehouse infrastructure for Hadoop that has an SQL-like interface.
