“An Introduction to MapReduce” by Pete Warden; O’Reilly Media

Founder of OpenHeatMap Pete Warden provides a simple introduction to MapReduce, the Google-created framework for examining large datasets. This video tutorial includes four videos: an intro to the topic, writing your first MapReduce job, and then two videos on running MapReduce jobs on the Amazon Elastic MapReduce cloud service. By the time you’re done watching the videos, you’ll understand what MapReduce is and how to write simple MapReduce jobs, and you’ll be ready to move on to more advanced MapReduce topics.

The videos use Python as the language for writing your mapper and reducer, but the code stays simple and even programmers unfamiliar with Python should follow along just fine. The tutorial has you writing real code, working on real data, and producing real results. You learn the principles necessary to apply to more advanced mapping and reducing scenarios. You can reasonably launch from the course material to write complex MapReduce jobs, depending on your programming skills and imagination.

This is definitely an introduction; the material moves a little slow, and some of the questions from the audience are pretty basic. Part of the Amazon foray is troubleshooting some of the audience members’ setups, for example, though Warden’s lead is straightforward throughout. For novice programmers, I’m sure the troubleshooting scenes are valuable; for professional programmers, they can be a bit tedious. It’s a small price to pay, however, for the information you’ll extract.

If you’re looking to take the mystery out of MapReduce and understand how to use the Amazon Elastic MapReduce service to run your own MapReduce jobs, you’ll find what you need here.