键值对 RDD
val rdd = sc.parallelize(Array(("spark",2),("hadoop",6),("hadoop",4),("spark",6)))
按key 分组求平均值
结果  ("spark",4),("hadoop",5)


方法1
val rdd = sc.parallelize(Array(("spark",2),("hadoop",6),("hadoop",4),("spark",6)))
val rdd2 = rdd.groupByKey()
rdd2.collect
//Array[(String, Iterable[Int])] = Array((spark,CompactBuffer(2, 6)), (hadoop,CompactBuffer(6, 4)))
rdd2.mapValues(v=>v.sum/v.size).collect
Array[(String, Int)] = Array((spark,4), (hadoop,5))

方法2
val rdd = sc.parallelize(Array(("spark",2),("hadoop",6),("hadoop",4),("spark",6)))
val rdd2 = rdd.groupByKey()
rdd2.collect
//Array[(String, Iterable[Int])] = Array((spark,CompactBuffer(2, 6)), (hadoop,CompactBuffer(6, 4)))

val rdd3 = rdd2.map(t=>(t._1,t._2.sum /t._2.size))
rdd3.collect
//Array[(String, Int)] = Array((spark,4), (hadoop,5))