My dear friend Mario has finally got around to blogging and one of his latest posts shows , it is possible to run Hadoop (which if you are not familiar with, can be thought of as an open source version of Google’s MapReduce ) in Azure. You need to setup a typical configuration of nodes (Name Nodes, Tracker and Slaves).
Sure, there are a number of dependencies some you would expect, others not (e.g. Cygwin – cringe!; but hopefully that will go away with Hadoop 0.22). I wonder what overheard the Cygwin runtime has?
It would be interesting to know if someone is (or planning) to use this at work.
