Created on 04-27-2016 06:57 PM
Demonstrate how SparkSQL can act as a distributed data federation platform.
Tables are created from three different sources:
Spark makes all of these tables available via JDBC as if from as single data store and speeds up processing by caching tables in memory.