Honestly, I remember seeing the announcement, but I'm not sure if the documentation has been updated, or if a guide has been produced. The only document that was released was the announcement itself, pointing you to the Apache Hadoop Project. I suspect that they anticipate that whomever will use it will already be familiar with HDFS, but that's not necessarily a fair assumption.
You may want to open a ticket with IBM, and request documentation on setting up CMOD to be used with Hadoop. I'll see if I can find someone to do a tutorial.
-JD.