Spark中aggregateByKey实现reduceByKey与groupByKey

Spark中aggregateByKey实现reduceByKey与groupByKey，下面为具体函数

  def aggregateByKeyToReduceBy(sc:SparkContext): Unit ={
    val rdd = sc.parallelize(List(("a",1),("b",2),("c",3),("a",2),("a",3),("b",3),("c",2)))
    rdd.reduceByKey(_+_).foreach(println)
    println("上面为reduceByKey的运算结果，下面为使用aggregateByKey算子的运算结果")

    val reduceByKeyRdd: RDD[(String, Int)] = rdd.aggregateByKey(0)(
      seqOp1,
      combOp1
    )
    reduceByKeyRdd.foreach(println)

  }
  //这里的参数arrayBuffer为rdd的value值的列表，v为rdd的value，这里为相同分区的聚合
  def seqOp1(c:Int,v:Int): Int ={
    c+v
  }
  //类似于上面那个函数，这里为不同分区相同key值value的聚合
  def combOp1(c1:Int,c2:Int): Int ={
    c1+c2
  }

  
  def aggregateByKeyToGroupBy(sc:SparkContext): Unit ={
    val rdd = sc.parallelize(List(("a",1),("b",2),("c",3),("a",2),("a",3),("b",3),("c",2)))
    rdd.groupByKey().foreach(println)
    println("上面为用groupByKey效果，下面为使用aggregateByKey算子实现的效果")

    val groupByKeyRdd: RDD[(String, ArrayBuffer[String])] = rdd.aggregateByKey(ArrayBuffer[String]())(
      seqOp,
      combOp
    )
    groupByKeyRdd.foreach(println)
  }

  //这里的参数arrayBuffer为rdd的value值的列表，v为rdd的value，这里为相同分区的聚合
  def seqOp(arrayBuffer: ArrayBuffer[String],v:Int): ArrayBuffer[String] ={
    arrayBuffer.append(v.toString)
    arrayBuffer
  }
  //类似于上面那个函数，这里为不同分区相同key值value的聚合
  def combOp(arrayBuffer1: ArrayBuffer[String],arrayBuffer2: ArrayBuffer[String]): ArrayBuffer[String] ={
    arrayBuffer1 ++= (arrayBuffer2)
  }