参数为1个RDD的Cogroup

import org.apache.spark.SparkContext
import org.apache.spark.SparkConf


object Cogroup {
  def main(args: Array[String]): Unit = {
    val sc = new SparkContext("local", "Cogroup", new SparkConf())


    cogroupTrans(sc)


    sc.stop()
  }
  def cogroupTrans(sc: SparkContext): Unit = {
    val stuNames = Array(
      Tuple2(1, "Spark"),
      Tuple2(2, "Tecc"),
      Tuple2(3, "Hadoop"))
    val stuScores = Array(
      Tuple2(1, 100),
      Tuple2(1, 99),
      Tuple2(2, 95),
      Tuple2(3, 65))
    val names = sc.parallelize(stuNames)
    val scores = sc.parallelize(stuScores)
    val stuNameAndScore = names.cogroup(scores)
    stuNameAndScore.collect().foreach(println)
  }

}

运行结果：

(1,(CompactBuffer(Spark),CompactBuffer(100, 99)))
(3,(CompactBuffer(Hadoop),CompactBuffer(65)))
(2,(CompactBuffer(Tecc),CompactBuffer(95)))

两个RDD先各自按照key进行分组，再进行联合分组