Statistical Computation with Incanter and MongoDB
Q: Can you explain why is MongoDB a good choice for incanter (as opposed to Clojure more generally?) Everything that I work with (in R) that is not rectangular, is indexed …
A: Do you mean as opposed to SQL databases, or other schema-less databases?
MongoDB can easily persist arbitrarily deeply nested Clojure data structures, which makes it a convenient choice, but that’s not to say there are not many other options, all equally useful.
I’d speculate that for larger data sets, having Incanter to work with HBase or Hadoop (if it doesn’t already) would take it to the next level.
