examples/src/main/scala/spark/examples/SkewedGroupByTest.scala


1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

package spark.examples

import spark.SparkContext
import spark.SparkContext._
import java.util.Random

object SkewedGroupByTest {
  def main(args: Array[String]) {
    if (args.length == 0) {
      System.err.println("Usage: GroupByTest <host> [numMappers] [numKVPairs] [KeySize] [numReducers]")
      System.exit(1)
    }  
    
    var numMappers = if (args.length > 1) args(1).toInt else 2
    var numKVPairs = if (args.length > 2) args(2).toInt else 1000
    var valSize = if (args.length > 3) args(3).toInt else 1000
    var numReducers = if (args.length > 4) args(4).toInt else numMappers

    val sc = new SparkContext(args(0), "GroupBy Test")
    
    val pairs1 = sc.parallelize(0 until numMappers, numMappers).flatMap { p =>
      val ranGen = new Random

      // map output sizes lineraly increase from the 1st to the last
      numKVPairs = (1. * (p + 1) / numMappers * numKVPairs).toInt

      var arr1 = new Array[(Int, Array[Byte])](numKVPairs)
      for (i <- 0 until numKVPairs) {
        val byteArr = new Array[Byte](valSize)
        ranGen.nextBytes(byteArr)
        arr1(i) = (ranGen.nextInt(Int.MaxValue), byteArr)
      }
      arr1
    }.cache
    // Enforce that everything has been calculated and in cache
    pairs1.count
    
    println(pairs1.groupByKey(numReducers).count)

    System.exit(0)
  }
}