abstract class RDD[T: ClassManifest](
    @transient private var sc: SparkContext, //@transient, 不需要序列化
    @transient private var deps: Seq[Dependency[_]]
  ) extends Serializable with Logging {
  

  /**辅助构造函数, 专门用于初始化1对1依赖关系的RDD,这种还是很多的, filter, map... 


Construct an RDD with just a one-to-one dependency on one parent */
  def this(@transient oneParent: RDD[_]) = 
    this(oneParent.context , List(new OneToOneDependency(oneParent)))  // 不同于一般的RDD, 这种情况因为只有一个parent, 所以直接传入parent RDD对象即可

  // =======================================================================
  // Methods that should be implemented by subclasses of RDD
  // =======================================================================
  /** Implemented by subclasses to compute a given partition. */
  def compute(split: Partition, context: TaskContext): Iterator[T]

  /**
   * Implemented by subclasses to return the set of partitions in this RDD. This method will only
   * be called once, so it is safe to implement a time-consuming computation in it.
   */
  protected def getPartitions: Array[Partition]

  /**
   * Implemented by subclasses to return how this RDD depends on parent RDDs. This method will only
   * be called once, so it is safe to implement a time-consuming computation in it.
   */
  protected def getDependencies: Seq[Dependency[_]] = deps

  /** Optionally overridden by subclasses to specify placement preferences. */
  protected def getPreferredLocations(split: Partition): Seq[String] = Nil

  /** Optionally overridden by subclasses to specify how they are partitioned. */
  val partitioner: Option[Partitioner] = None

  // =======================================================================
  // Methods and fields available on all RDDs
  // =======================================================================

  /** The SparkContext that created this RDD. */
  def sparkContext: SparkContext = sc

  /** A unique ID for this RDD (within its SparkContext). */
  val id: Int = sc.newRddId()

  /** A friendly name for this RDD */
  var name: String = null

  /**
   * Set this RDD's storage level to persist its values across operations after the first time
   * it is computed. This can only be used to assign a new storage level if the RDD does not
   * have a storage level set yet..
   */
  def persist(newLevel: StorageLevel): RDD[T] = {
    // TODO: Handle changes of StorageLevel
    if (storageLevel != StorageLevel.NONE && newLevel != storageLevel) {
      throw new UnsupportedOperationException(
        "Cannot change storage level of an RDD after it was already assigned a level")
    }
    storageLevel = newLevel
    // Register the RDD with the SparkContext
    sc.persistentRdds(id) = this
    this
  }

  /** Persist this RDD with the default storage level (`MEMORY_ONLY`). */
  def persist(): RDD[T] = persist(StorageLevel.MEMORY_ONLY)

  /** Persist this RDD with the default storage level (`MEMORY_ONLY`). */
  def cache(): RDD[T] = persist()

  /** Get the RDD's current storage level, or StorageLevel.NONE if none is set. */
  def getStorageLevel = storageLevel

  // Our dependencies and partitions will be gotten by calling subclass's methods below, and will
  // be overwritten when we're checkpointed
  private var dependencies_ : Seq[Dependency[_]] = null  
  @transient private var partitions_ : Array[Partition] = null 

  /** An Option holding our checkpoint RDD, if we are checkpointed 

* checkpoint就是把RDD存到磁盘文件中, 以提高failover的效率, 虽然也可以选择replay

* 并且在RDD的实现中, 如果存在checkpointRDD, 则可以直接从中读到RDD数据, 而不需要compute */
  private def checkpointRDD: Option[RDD[T]] = checkpointData.flatMap(_.checkpointRDD)

  /**
   * Internal method to this RDD; will read from cache if applicable, or otherwise compute it.
   * This should ''not'' be called by users directly, but is available for implementors of custom
   * subclasses of RDD.
   */

  /** 这是RDD访问数据的核心, 在RDD中的Partition中只包含id而没有真正数据

* 那么如果获取RDD的数据? 参考storage模块 

* 在cacheManager.getOrCompute中, 会将RDD和Partition id对应到相应的block, 并从中读出数据*/ 
  final def iterator(split: Partition, context: TaskContext): Iterator[T] = {
    if (storageLevel != StorageLevel.NONE) {//StorageLevel不为None,说明这个RDD persist过, 可以直接读出来
      SparkEnv.get.cacheManager.getOrCompute(this, split, context, storageLevel)
    } else {
      computeOrReadCheckpoint(split, context) //如果没有persisit过, 只有从新计算出, 或从checkpoint中读出
    }
  }

  // Transformations (return a new RDD)
  //...... 各种transformations的接口,map, union...

  /**
   * Return a new RDD by applying a function to all elements of this RDD.
   */
  def map[U: ClassManifest](f: T => U): RDD[U] = new MappedRDD(this, sc.clean(f))

  // Actions (launch a job to return a value to the user program)
  //......各种actions的接口,count, collect...

  /**
   * Return the number of elements in the RDD.
   */
  def count(): Long = {// 只有在action中才会真正调用runJob, 所以transform都是lazy的
    sc.runJob(this, (iter: Iterator[T]) => {
      var result = 0L
      while (iter.hasNext) {
        result += 1L
        iter.next()
      }
      result
    }).sum
  }

  // =======================================================================
  // Other internal methods and fields
  // =======================================================================  

  /** Returns the first parent RDD 

返回第一个parent RDD*/
  protected[spark] def firstParent[U: ClassManifest] = {
    dependencies.head.rdd.asInstanceOf[RDD[U]]
  }

  //................
}

  /**
   * Return a new RDD containing only the elements that satisfy a predicate.
   */
  def filter(f: T => Boolean): RDD[T] = new FilteredRDD(this, sc.clean(f))

private[spark] class FilteredRDD[T: ClassManifest]( //filter是典型的one-to-one dependency, 使用辅助构造函数 
    prev: RDD[T],   //parent RDD
    f: T => Boolean) //f,过滤函数
  extends RDD[T](prev) {
  //firstParent会从deps中取出第一个RDD对象, 就是传入的prev RDD, 在One-to-one Dependency中,parent和child的partition信息相同
  override def getPartitions: Array[Partition] = firstParent[T].partitions

  override val partitioner = prev.partitioner    // Since filter cannot change a partition's keys

  override def compute(split: Partition, context: TaskContext) =
    firstParent[T].iterator(split, context).filter(f) //compute就是真正产生RDD的逻辑
}

  /**
   * Return the union of this RDD and another one. Any identical elements will appear multiple
   * times (use `.distinct()` to eliminate them).
   */
  def union(other: RDD[T]): RDD[T] = new UnionRDD(sc, Array(this, other))

private[spark] class UnionPartition[T: ClassManifest](idx: Int, rdd: RDD[T], splitIndex: Int)
  extends Partition {

  var split: Partition = rdd.partitions(splitIndex)//从parent RDD中取出相应的partition, 重用

  def iterator(context: TaskContext) = rdd.iterator(split, context)//Iterator也可以重用

  def preferredLocations() = rdd.preferredLocations(split)

  override val index: Int = idx//partition id是新的, 因为多个合并后, 序号肯定会发生变化
}

class UnionRDD[T: ClassManifest](
    sc: SparkContext,
    @transient var rdds: Seq[RDD[T]])  //parent RDD Seq
  extends RDD[T](sc, Nil) {  // Nil since we implement getDependencies

  override def getPartitions: Array[Partition] = {
    val array = new Array[Partition](rdds.map(_.partitions.size).sum) //UnionRDD的partition数,是所有parent RDD中的partition数目的和
    var pos = 0
    for (rdd <- rdds; split <- rdd.partitions) {
      array(pos) = new UnionPartition(pos, rdd, split.index) //创建所有的UnionPartition
      pos += 1
    }
    array
  }

  override def getDependencies: Seq[Dependency[_]] = {
    val deps = new ArrayBuffer[Dependency[_]]
    var pos = 0
    for (rdd <- rdds) { 
      deps += new RangeDependency(rdd, 0, pos, rdd.partitions.size)//创建RangeDependency
      pos += rdd.partitions.size)//由于是RangeDependency, 所以pos的递增是加上整个区间size
    }
    deps
  }

  override def compute(s: Partition, context: TaskContext): Iterator[T] =
    s.asInstanceOf[UnionPartition[T]].iterator(context)//Union的compute非常简单,什么都不需要做

  override def getPreferredLocations(s: Partition): Seq[String] =
    s.asInstanceOf[UnionPartition[T]].preferredLocations()
}

本文章摘自博客园，原文发布日期：2013-12-24