学习进度笔记

学习进度笔记31

PageRank 演示

import org.apache.log4j.{Level, Logger}

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

import org.apache.spark.graphx._

import org.apache.spark.rdd.RDD

 

object PageRank {

  def main(args: Array[String]) {

    //屏蔽日志

    Logger.getLogger("org.apache.spark").setLevel(Level.WARN)

    Logger.getLogger("org.eclipse.jetty.server").setLevel(Level.OFF)

 

    //设置运行环境

    val conf = new SparkConf().setAppName("PageRank").setMaster("local")

    val sc = new SparkContext(conf)

 

    //读入数据文件

    val articles: RDD[String] = sc.textFile("/home/hadoop/IdeaProjects/data/graphx/graphx-wiki-vertices.txt")

    val links: RDD[String] = sc.textFile("/home/hadoop/IdeaProjects/data/graphx/graphx-wiki-edges.txt")

 

    //装载顶点和边

    val vertices = articles.map { line =>

      val fields = line.split('\t')

      (fields(0).toLong, fields(1))

    }

 

    val edges = links.map { line =>

      val fields = line.split('\t')

      Edge(fields(0).toLong, fields(1).toLong, 0)

    }

 

    //cache操作

    //val graph = Graph(vertices, edges, "").persist(StorageLevel.MEMORY_ONLY_SER)

    val graph = Graph(vertices, edges, "").persist()

    //graph.unpersistVertices(false)

 

    //测试

    println("**********************************************************")

    println("获取5个triplet信息")

    println("**********************************************************")

    graph.triplets.take(5).foreach(println(_))

 

    //pageRank算法里面的时候使用了cache()，故前面persist的时候只能使用MEMORY_ONLY

    println("**********************************************************")

    println("PageRank计算，获取最有价值的数据")

    println("**********************************************************")

    val prGraph = graph.pageRank(0.001).cache()

 

    val titleAndPrGraph = graph.outerJoinVertices(prGraph.vertices) {

      (v, title, rank) => (rank.getOrElse(0.0), title)

    }

 

    titleAndPrGraph.vertices.top(10) {

      Ordering.by((entry: (VertexId, (Double, String))) => entry._2._1)

    }.foreach(t => println(t._2._2 + ": " + t._2._1))

 

    sc.stop()

  }

}