Cassandra .6 map/reduce

[Sonny Heer]

Running the word_count example the hadoop job appears to be run internally.

If I have a Cassandra cluster of 10 nodes, how does the Hadoop cluster
get configured?