I think we're running out of good names for things, especially since Spark uses Snappy (compression) itself.

It is time to build a tool on top of SnappyData called Spark.