For union all operator, we will use Spark's union transformation. Refer to the design doc on wiki for more information.
Investigate why union two RDDs generated from two MapTrans does not get the right result
Change SparkCompiler to generate a SparkWork that contains UnionWork from logical operator tree
Implement the graph transformation execution
Generate plan for spark work which uses spark union transformation
Add .q tests coverage for "union all" [Spark Branch]
hive.optimize.union.remove does not work properly [Spark Branch]