[christkv]

there's an initial hadoop plugin for mongo that might be a better fit for doing map-reduce over large datasets https://github.com/mongodb/mongo-hadoop