HDFS Persistence for Bolt Array RDD

[SteveAIS]

It would be nice to be able to persist to HDFS the RDD that is created through a load images call when a Spark context is passed. That would allow quick access later to that RDD without having to go through the load images call again. It could be stored with a call like sc.pickleFile(...)