Spark测试代码求PI的原理

Spark测试代码求PI的原理

原理图

代码实现

package com.core.day2

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

import scala.util.Random

object Demo20PI {
  def main(args: Array[String]): Unit = {
    val conf = new SparkConf()
    conf.setAppName("submit")

    conf.setMaster("local")

    val sc = new SparkContext(conf)

    //生成一个很大的集合
    val list: Range.Inclusive = 0 to 10000

    //构建一个很大的额rdd
    val rdd: RDD[Int] = sc.parallelize(list)

    //随机生成正方形内的点
    val pointRDD: RDD[(Double, Double)] = rdd.map(i => {
      //随机生成x和y,范围是-1到1
      val x: Double = Random.nextDouble() * 2 - 1
      val y: Double = Random.nextDouble() * 2 - 1
      (x, y)
    })

    //取出圆内的点
    val yuanRDD: RDD[(Double, Double)] = pointRDD.filter {
      case (x: Double, y: Double) =>
        x * x + y * y < 1
    }


    //计算Pi
    val PI: Double = 4.0 * yuanRDD.count() / pointRDD.count()

    println(PI)

  }

}