RDD key/value关联操作
val left = sc.parallelize(List(("spark",1),("hadoop",1),("storm",1)))
val left = sc.parallelize(List(("scala",1),("hadoop",1),("spark",1)))
关联2个RDD
val joinOut = left join right
res9: Array[(String, (Int, Int))] = Array((spark,(1,1)), (hadoop,(1,1)))
left.join(right)
(left cogroup right).collect