spark graphx图计算

graph

import org.apache.spark.graphx.{Edge, Graph, VertexId} import org.apache.spark.rdd.RDD import org.apache.spark.{SparkConf, SparkContext} //求共同好友 object CommendFriend {    def main(args: Array[String]): Unit = {     //创建入口     val conf: SparkConf = new SparkConf().setAppName("CommendFriend").setMaster("local[*]")     val sc: SparkContext = new SparkContext(conf)     //点的集合     //点     val uv: RDD[(VertexId,(String,Int))] = sc.parallelize(Seq(       (133, ("毕东旭", 58)),       (1, ("贺咪咪", 18)),       (2, ("范闯", 19)),       (9, ("贾璐燕", 24)),       (6, ("马彪", 23)),        (138, ("刘国建", 40)),       (16, ("李亚茹", 18)),       (21, ("任伟", 25)),       (44, ("张冲霄", 22)),        (158, ("郭佳瑞", 22)),       (5, ("申志宇", 22)),       (7, ("卫国强", 22))     ))     //边的集合     //边Edge     val ue: RDD[Edge[Int]] = sc.parallelize(Seq(       Edge(1, 133,0),       Edge(2, 133,0),       Edge(9, 133,0),       Edge(6, 133,0),        Edge(6, 138,0),       Edge(16, 138,0),       Edge(44, 138,0),       Edge(21, 138,0),        Edge(5, 158,0),       Edge(7, 158,0)     ))     //构建图（连通图）     val graph: Graph[(String, Int), Int] = Graph(uv,ue)     //调用连通图算法     graph       .connectedComponents()       .vertices       .join(uv)       .map{         case (uid,(minid,(name,age)))=>(minid,(uid,name,age))       }.groupByKey()       .foreach(println(_))     //关闭   } }

二、用户标签数据合并Demo

测试数据

陌上花开 旧事酒浓 多情汉子 APP:10 BS:8 多情汉子 满心闯 K:20 满心闯 喜欢不是爱 不是唯一 APP:10 装逼卖萌无所不能 K:5

计算结果数据

(-397860375,(List(, , , , , , , ),List((APP,20), (K,20), (BS,8)))) (553023549,(List(),List((K,5))))

来源：博客园

作者：lilixia

链接：https://www.cnblogs.com/JBLi/p/11552443.html