The scalable JSON based document storage engine.
Mongo DB is a highly scalable, NoSQL, JSON based, schema free data store. It's designed to readily scale from small to medium, to very large. It's also jolly nice to work with.
The aggregate pipeline allows you to compose complex queries which execute across large data sets surprisingly quickly, even without indexes. You can pull out portions of the data set, or you can manipulate the data set and represent it in any number of ways using expressions.
Built in Map Reduce allows you to drop right into code should you come across a problem that the aggregate pipeline can't handle.