Last year I posted about a project to build a Map-Reduce-like Windows Azure High-Scale compute architecture as part of the work I did on a project for a major Pharmaceutical company. 

image

The major Java Map-Reduce framework in use today is Apache Hadoop.  Originally patterned after the proprietary Google Map-Reduce framework, it is currently Open Source and implemented on a number of platforms.  

Now we have a Map-Reduce solution for Windows Azure.  This blog post describes how to set it up.

The blog post demonstrates in detail how to create a typical Hadoop cluster with a Name Node, a Job Tracker and a customizable number of Slaves.