MSSP

run API

• API

  def run[VD, ED: ClassTag](graph: Graph[VD, ED], landmarks: Seq[VertexId]): Graph[SPMap, ED]

• 功能描述

  计算图中所有节点到指定节点的最短路径距离。

• API描述
  1. 包名：package org.apache.spark.graphx.lib.ShortestPaths
  2. 类名：ShortestPaths
  3. 方法名：run
  4. 输入：
     1. graph: Graph[VD,ED]
     2. landmarks: Seq[vertexId]
  5. 参数详情：

     参数名称	参数含义	取值类型
     graph	从文件中读入后的Graph数据	Graph[VD,ED]
     landmarks	指定节点序列	Seq[VertexId]

  6. 输出：resultGraph:Graph[SPMap,ED]
• 使用样例

  MSSP样例：

  val sc = new SparkContext(new SparkConf().setMaster("yarn").setAppName("MSSP"))
  val inputGraphRaw = Array((1L, 2L), (3L, 2L), (3L, 1L), (2L, 3L), (4L, 2L) , (5L, 6L))
  val sourceRaw = Array(1L,2L)
  val inputGraph = sc.parallelize(inputGraphRaw)
  val graph = GraphDataLoader.convertToGraphXGraph(inputGraphRaw).persist()
  val source = sc.parallelize(sourceRaw)
  val res = run(graph, sources.collect.toSeq).vertices.collectAsMap()

• 样例结果

  Res:
  Map(
  2 -> Map(2 -> 0, 1 -> 2),
  4 -> Map(2 -> 1, 1 -> 3), 1 -> Map(1 -> 0, 2 -> 1),
  3 -> Map(1 -> 1, 2 -> 1),
  5-> Map(),
  6-> Map())