Web-Scale Information Processing Applications

An Integrated Research and Educational Initiative with MapReduce

(Spring 2008)

Jimmy Lin

Note: This course has concluded and this site is no longer being actively maintained.

Google is currently rumored to have approximately 450,000 active machines... ever wonder how one might write a program to exploit that much processing power? The answer is a powerful programming paradigm called MapReduce. Fortunately for non-Googlers, there's an open source implementation called Hadoop. Thanks to a collaboration with IBM and Google, the University of Maryland has access to a Hadoop cluster.

The basic idea behind this course is to create small teams focused on open research problems, to explore how MapReduce can be used in interesting and novel ways. The teams will consist of Ph.D. students ("team leaders") and both masters and undergraduate students ("team members"). See the projects page for a run-down of what we're working on. Also of interest is the cloud computing speaker series, where we'll bring in researchers to talk about related work.

News!

Nothing currently, but there's always this.

Older news...

Course Information

This page, first created: 17 Oct 2007; last updated: Creative Commons: Attribution-Noncommercial-Share Alike 3.0 United States Valid XHTML 1.0! Valid CSS!