由于天然符合互联网中很多场景的需求，图计算正受到越来越多的青睐。Spark GraphX 是作为 Spark 技术堆栈中的一员，担负起了 Spark 在图计算领域中的重任。网络上已经有很多图计算和 Spark GraphX 的概念介绍，此处就不再赘述。 本文将一篇很好的 Spark GraphX 入门文章中代码块整合为一个完整的可执行类，并加上必要注释以及执行结果，以方便有兴趣的朋友快速从 API 角度了解 Spark GraphX。
本文引用的代码块和多数文字描述均摘引自网文 graph-analytics-with-graphx， 在此特向作者表以感谢！
[1] 完整可执行scala 代码：
package scala.spark.graphx
import org.apache.spark.graphx._
import org.apache.spark.rdd.RDD
import org.apache.spark._
import org.apache.spark.SparkContext._
object GraphXExample {
def main(args: Array[String]) {
val conf = new SparkConf().setAppName("GraphXExample")
val sc = new SparkContext(conf)
// [A] creating the Property Graph from arrays of vertices and edges
println("[A] creating the Property Graph from arrays of vertices and edges");
// Each vertex is keyed by a unique 64-bit long identifier (VertexID), like '1L'
val vertexArray = Array(
(1L, ("Alice", 28)),
(2L, ("Bob", 27)),
(3L, ("Charlie", 65)),
(4L, ("David", 42)),
(5L, ("Ed", 55)),
(6L, ("Fran", 50)))
// the Edge class stores a srcId, a dstId and the edge property
val edgeArray = Array(
Edge(2L, 1L, 7),
Edge(2L, 4L, 2),
Edge(3L, 2L, 4),
Edge(3L, 6L, 3),
Edge(4L, 1L, 1),
Edge(5L, 2L, 2),
Edge(5L, 3L, 8),
Edge(5L, 6L, 3))
// construct the following RDDs from the vertexArray and edgeArray variables.
val vertexRDD: RDD[(Long, (String, Int))] = sc.parallelize(vertexArray)
val edgeRDD: RDD[Edge[Int]] = sc.parallelize(edgeArray)
// build a Property Graph
val graph: Graph[(String, Int), Int] = Graph(vertexRDD, edgeRDD)
// [B] Extract the vertex and edge RDD views of a graph
println("[B] Extract the vertex and edge RDD views of a graph");
// Solution 1
println("Solution 1:============")
graph.vertices.filter { case (id, (name, age)) => age > 30 }.collect.foreach {
case (id, (name, age)) => println(s"$name is $age")
}
// Solution 2
println("Solution 2:============")
graph.vertices.filter(v => v._2._2 > 30).collect.foreach(v => println(s"${v._2._1} is ${v._2._2}"))
// Solution 3
println("Solution 3:============")
for ((id, (name, age)) <- graph.vertices.filter { case (id, (name, age)) => age > 30 }.collect) {
println(s"$name is $age")
}
// [C] Exposes a triplet view which logically joins the vertex and edge properties yielding an RDD[EdgeTriplet[VD, ED]]
println("[C] Exposes a triplet view which logically joins the vertex and edge properties yielding an RDD[EdgeTriplet[VD, ED]]");
println("Use the graph.triplets view to display who likes who: ")
for (triplet <- graph.triplets.collect) {
println(s"${triplet.srcAttr._1} likes ${triplet.dstAttr._1}")
}
// For extra credit, find the lovers.
// If someone likes someone else more than 5 times than that relationship is getting pretty serious.
println("For extra credit, find the lovers if has:============")
for (triplet <- graph.triplets.filter(t => t.attr > 5).collect) {
println(s"${triplet.srcAttr._1} loves ${triplet.dstAttr._1}")
}
// [D] Graph Operators
// Property Graphs also have a collection of basic operations
println("[D] Graph Operators")
// compute the in-degree of each vertex
val inDegrees: VertexRDD[Int] = graph.inDegrees
// Define a class to more clearly model the user property
case class User(name: String, age: Int, inDeg: Int, outDeg: Int)
// Create a user Graph
val initialUserGraph: Graph[User, Int] = graph.mapVertices { case (id, (name, age)) => User(name, age, 0, 0) }
// Fill in the degree information
val userGraph = initialUserGraph.outerJoinVertices(initialUserGraph.inDegrees) {
case (id, u, inDegOpt) => User(u.name, u.age, inDegOpt.getOrElse(0), u.outDeg)
}.outerJoinVertices(initialUserGraph.outDegrees) {
case (id, u, outDegOpt) => User(u.name, u.age, u.inDeg, outDegOpt.getOrElse(0))
}
// Here we use the outerJoinVertices method of Graph which has the following (confusing) type signature:
// def outerJoinVertices[U, VD2](other: RDD[(VertexID, U)])(mapFunc: (VertexID, VD, Option[U]) => VD2): Graph[VD2, ED]
// Using the degreeGraph print the number of people who like each user:
println("Using the degreeGraph print the number of people who like each user:============")
for ((id, property) <- userGraph.vertices.collect) {
println(s"User $id is called ${property.name} and is liked by ${property.inDeg} people.")
}
// Print the names of the users who are liked by the same number of people they like.
userGraph.vertices.filter {
case (id, u) => u.inDeg == u.outDeg
}.collect.foreach {
case (id, property) => println(property.name)
}
// [D.1] The Map Reduce Triplets Operator
// The mapReduceTriplets operator enables neighborhood aggregation and find the oldest follower of each user
println("[D.1] The Map Reduce Triplets Operator")
// Find the oldest follower for each user
println("Find the oldest follower for each user:============")
val oldestFollower: VertexRDD[(String, Int)] = userGraph.mapReduceTriplets[(String, Int)](
// For each edge send a message to the destination vertex with the attribute of the source vertex
edge => Iterator((edge.dstId, (edge.srcAttr.name, edge.srcAttr.age))),
// To combine messages take the message for the older follower
(a, b) => if (a._2 > b._2) a else b)
userGraph.vertices.leftJoin(oldestFollower) { (id, user, optOldestFollower) =>
optOldestFollower match {
case None => s"${user.name} does not have any followers."
case Some((name, age)) => s"${name} is the oldest follower of ${user.name}."
}
}.collect.foreach { case (id, str) => println(str) }
// Try finding the average follower age of the followers of each user
println("Try finding the average follower age of the followers of each user:============")
val averageAge: VertexRDD[Double] = userGraph.mapReduceTriplets[(Int, Double)](
// map function returns a tuple of (1, Age)
edge => Iterator((edge.dstId, (1, edge.srcAttr.age.toDouble))),
// reduce function combines (sumOfFollowers, sumOfAge)
(a, b) => ((a._1 + b._1), (a._2 + b._2))).mapValues((id, p) => p._2 / p._1)
// Display the results
userGraph.vertices.leftJoin(averageAge) { (id, user, optAverageAge) =>
optAverageAge match {
case None => s"${user.name} does not have any followers."
case Some(avgAge) => s"The average age of ${user.name}\'s followers is $avgAge."
}
}.collect.foreach { case (id, str) => println(str) }
// [D.2] Subgraph
// The subgraph operator that takes vertex and edge predicates and returns the graph
// containing only the vertices that satisfy the vertex predicate (evaluate to true)
// and edges that satisfy the edge predicate and connect vertices that satisfy the
// vertex predicate.
println("[D.2] Subgraph")
// restrict our graph to the users that are 30 or older
println("restrict our graph to the users that are 30 or older:============")
val olderGraph = userGraph.subgraph(vpred = (id, user) => user.age >= 30)
// compute the connected components
val cc = olderGraph.connectedComponents
// display the component id of each user:
olderGraph.vertices.leftJoin(cc.vertices) {
case (id, user, comp) => s"${user.name} is in component ${comp.get}"
}.collect.foreach { case (id, str) => println(str) }
}
}[2] 执行结果：
[A] creating the Property Graph from arrays of vertices and edges [B] Extract the vertex and edge RDD views of a graph Solution 1:============ David is 42 Fran is 50 Charlie is 65 Ed is 55 Solution 2:============ David is 42 Fran is 50 Charlie is 65 Ed is 55 Solution 3:============ David is 42 Fran is 50 Charlie is 65 Ed is 55 [C] Exposes a triplet view which logically joins the vertex and edge properties yielding an RDD[EdgeTriplet[VD, ED]] Use the graph.triplets view to display who likes who: Bob likes Alice Bob likes David Charlie likes Bob Charlie likes Fran David likes Alice Ed likes Bob Ed likes Charlie Ed likes Fran For extra credit, find the lovers if has:============ Bob loves Alice Ed loves Charlie [D] Graph Operators Using the degreeGraph print the number of people who like each user:============ User 4 is called David and is liked by 1 people. User 6 is called Fran and is liked by 2 people. User 2 is called Bob and is liked by 2 people. User 1 is called Alice and is liked by 2 people. User 3 is called Charlie and is liked by 1 people. User 5 is called Ed and is liked by 0 people. David Bob [D.1] The Map Reduce Triplets Operator Find the oldest follower for each user:============ Bob is the oldest follower of David. Charlie is the oldest follower of Fran. Charlie is the oldest follower of Bob. David is the oldest follower of Alice. Ed is the oldest follower of Charlie. Ed does not have any followers. Try finding the average follower age of the followers of each user:============ The average age of David‘s followers is 27.0. The average age of Fran‘s followers is 60.0. The average age of Bob‘s followers is 60.0. The average age of Alice‘s followers is 34.5. The average age of Charlie‘s followers is 55.0. Ed does not have any followers. [D.2] Subgraph restrict our graph to the users that are 30 or older:============ David is in component 4 Fran is in component 3 Charlie is in component 3 Ed is in component 3 |
原文地址：http://blog.csdn.net/samhacker/article/details/42045539
