spark core api 的一些demo

1、hello world 
  val sc = SparkContext("Spark://localhost:7077","Hello world","SPARK_HOME","YOUR_APP_JAR")
  val file = sc.textFile("hdfs://")
  val filterRDD = file.filter(_contains("Hello World"))
  filterRDD.cache()
  filterRDD.count()

2、RDD分区
  val rdd = sc.parallelize(1 to 100,2)
  rdd.partitions.size
  val rdd = sc.parallelize(1 to 100)
  rdd.partitions.size

3、RDD优先位置
   val rdd = sc.textFile(hdfs:10.0.2.19:9000/bigfile)
   val hadoopRDD = rdd.dependencies(0).rdd
   hadoopRDD.partitions.size
   hadoopRDD.preferredLocations(hadoopRDD.partitions(0))

4、RDD的依赖关系
   val rdd = sc.makeRDD(1 to 10)
   val mapRDD = rdd.map(x => (x,x))
   mapRDD.dependencies
   val shuffleRDD = mapRDD.partitionBy(new org.apache.spark.HashPartitioner(3))   ？？？？？
   shuffleRDD.dependencies

5、PageRank
   val links = sc.parallelize(Array(('A',Array('D')),('B',Array('A')),('C',Array('A','B')),('D',Array('A','B'))),2)
                  .map(x =>(x._1,x._2)).cache()
   val ranks = sc.parallelize(Array(('A',1.0),('B',1.0),('C',1.0),('D',1.0)),2)