[Spark] 페어 RDD의 트랜스포메이션 예제

package com.ex

import org.apache.spark.{SparkConf, SparkContext}

/**
 * Created by moonsun on 17. 3. 14..
 */
object RddTest {

  def main(args: Array[String]): Unit = {
     val conf = new SparkConf().setMaster("local").setAppName("RDD TEST")
     val sc = new SparkContext(conf)

     val input = sc.parallelize(List((1,2),(3,6),(2,3),(2,5)))

     println("\ntt0")
     val tt0 = input.reduceByKey( (x,y)=> x+y)
     tt0.foreach{ println }

     println("\ntt1")
     val tt1 = input.groupByKey()
     tt1.foreach{ println }

     println("\ntt2")
     val tt2 = input.keys
     tt2.foreach{ println }

     println("\ntt3")
     val tt3 = input.values
     tt3.foreach{ println }

     println("\ntt4")
     val tt4 = input.mapValues(x => x + 1)
     tt4.foreach{ println }

     println("\ntt5")
     val tt5 = input.sortByKey()
     tt5.foreach{ println }

  }

}

==> 결과

tt0
(1,2)
(3,6)
(2,8)

tt1
(1,CompactBuffer(2))
(3,CompactBuffer(6))
(2,CompactBuffer(3, 5))

tt2
1
3
2
2

tt3
2
6
3
5

tt4
(1,3)
(3,7)
(2,4)
(2,6)

tt5
(1,2)
(2,3)
(2,5)
(3,6)


-- 끝 --