十、DAGScheduler内幕

一 runJob

上节介绍到调用DAGScheduler的runJob，DAGScheduler runJob调用submitJob提交job。

  /**
   * Run an action job on the given RDD and pass all the results to the resultHandler function as
   * they arrive.
   *
   * @param rdd target RDD to run tasks on
   * @param func a function to run on each partition of the RDD
   * @param partitions set of partitions to run on; some jobs may not want to compute on all
   *   partitions of the target RDD, e.g. for operations like first()
   * @param callSite where in the user program this job was called
   * @param resultHandler callback to pass each result to
   * @param properties scheduler properties to attach to this job, e.g. fair scheduler pool name
   *
   * @note Throws `Exception` when the job fails
   */
  def runJob[T, U](
      rdd: RDD[T],
      func: (TaskContext, Iterator[T]) => U,
      partitions: Seq[Int],
      callSite: CallSite,
      resultHandler: (Int, U) => Unit,
      properties: Properties): Unit = {
    val start = System.nanoTime
    val waiter = submitJob(rdd, func, partitions, callSite, resultHandler, properties) // 提交job
    ThreadUtils.awaitReady(waiter.completionFuture, Duration.Inf)
    waiter.completionFuture.value.get match {
      case scala.util.Success(_) =>
        logInfo("Job %d finished: %s, took %f s".format
          (waiter.jobId, callSite.shortForm, (System.nanoTime - start) / 1e9))
      case scala.util.Failure(exception) =>
        logInfo("Job %d failed: %s, took %f s".format
          (waiter.jobId, callSite.shortForm, (System.nanoTime - start) / 1e9))
        // SPARK-8644: Include user stack trace in exceptions coming from DAGScheduler.
        val callerStackTrace = Thread.currentThread().getStackTrace.tail
        exception.setStackTrace(exception.getStackTrace ++ callerStackTrace)
        throw exception
    }
  }

二 submitJob会向DAGSchedulerEventProcessLoop发送JobSubmitted事件。

  /**
   * Submit an action job to the scheduler.
   *
   * @param rdd target RDD to run tasks on
   * @param func a function to run on each partition of the RDD
   * @param partitions set of partitions to run on; some jobs may not want to compute on all
   *   partitions of the target RDD, e.g. for operations like first()
   * @param callSite where in the user program this job was called
   * @param resultHandler callback to pass each result to
   * @param properties scheduler properties to attach to this job, e.g. fair scheduler pool name
   *
   * @return a JobWaiter object that can be used to block until the job finishes executing
   *         or can be used to cancel the job.
   *
   * @throws IllegalArgumentException when partitions ids are illegal
   */
  def submitJob[T, U](
      rdd: RDD[T],
      func: (TaskContext, Iterator[T]) => U,
      partitions: Seq[Int],
      callSite: CallSite,
      resultHandler: (Int, U) => Unit,
      properties: Properties): JobWaiter[U] = {
    // Check to make sure we are not launching a task on a partition that does not exist.
    val maxPartitions = rdd.partitions.length
    partitions.find(p => p >= maxPartitions || p < 0).foreach { p =>
      throw new IllegalArgumentException(
        "Attempting to access a non-existent partition: " + p + ". " +
          "Total number of partitions: " + maxPartitions)
    }

    val jobId = nextJobId.getAndIncrement()
    if (partitions.size == 0) {
      // Return immediately if the job is running 0 tasks
      return new JobWaiter[U](this, jobId, 0, resultHandler)
    }

    assert(partitions.size > 0)
    val func2 = func.asInstanceOf[(TaskContext, Iterator[_]) => _]
    val waiter = new JobWaiter(this, jobId, partitions.size, resultHandler)
    // 发送JobSubmitted事件
    eventProcessLoop.post(JobSubmitted(
      jobId, rdd, func2, partitions.toArray, callSite, waiter,
      SerializationUtils.clone(properties)))
    waiter
  }

  private[scheduler] val eventProcessLoop = new DAGSchedulerEventProcessLoop(this)
  taskScheduler.setDAGScheduler(this)

三 DAGSchedulerEventProcessLoop处理消息。

private[scheduler] class DAGSchedulerEventProcessLoop(dagScheduler: DAGScheduler)
  extends EventLoop[DAGSchedulerEvent]("dag-scheduler-event-loop") with Logging {

  private[this] val timer = dagScheduler.metricsSource.messageProcessingTimer

  /**
   * The main event loop of the DAG scheduler.
   */
  override def onReceive(event: DAGSchedulerEvent): Unit = {
    val timerContext = timer.time()
    try {
      doOnReceive(event)
    } finally {
      timerContext.stop()
    }
  }

  private def doOnReceive(event: DAGSchedulerEvent): Unit = event match {
    // 处理JobSubmitted
    case JobSubmitted(jobId, rdd, func, partitions, callSite, listener, properties) =>
      dagScheduler.handleJobSubmitted(jobId, rdd, func, partitions, callSite, listener, properties)

...
  }

四 处理提交的job，调用submitStage提交stage。

  private[scheduler] def handleJobSubmitted(jobId: Int,
      finalRDD: RDD[_],
      func: (TaskContext, Iterator[_]) => _,
      partitions: Array[Int],
      callSite: CallSite,
      listener: JobListener,
      properties: Properties) {
    // 使用出发job的最后一个rdd创建finalStage，类型为ResultStage，因为最后一个stage为ResultStage
    var finalStage: ResultStage = null
    try {
      // New stage creation may throw an exception if, for example, jobs are run on a
      // HadoopRDD whose underlying HDFS files have been deleted.
      finalStage = createResultStage(finalRDD, func, partitions, jobId, callSite)
    } catch {
      case e: Exception =>
        logWarning("Creating new stage failed due to exception - job: " + jobId, e)
        listener.jobFailed(e)
        return
    }

    // 为finalStage创建一个job
    val job = new ActiveJob(jobId, finalStage, callSite, listener, properties)
    clearCacheLocs()
    logInfo("Got job %s (%s) with %d output partitions".format(
      job.jobId, callSite.shortForm, partitions.length))
    logInfo("Final stage: " + finalStage + " (" + finalStage.name + ")")
    logInfo("Parents of final stage: " + finalStage.parents)
    logInfo("Missing parents: " + getMissingParentStages(finalStage))

    val jobSubmissionTime = clock.getTimeMillis()
    jobIdToActiveJob(jobId) = job
    activeJobs += job
    finalStage.setActiveJob(job)
    val stageIds = jobIdToStageIds(jobId).toArray
    val stageInfos = stageIds.flatMap(id => stageIdToStage.get(id).map(_.latestInfo))
    listenerBus.post(
      SparkListenerJobStart(job.jobId, jobSubmissionTime, stageInfos, properties))
    submitStage(finalStage) //提交finalStage
  }

五 提交stage，调用getMissingParentStages获取父stage，进行stage划分，这是一个递归调用的过程。

  /** Submits stage, but first recursively submits any missing parents. */
  private def submitStage(stage: Stage) {
    val jobId = activeJobForStage(stage)
    if (jobId.isDefined) {
      logDebug("submitStage(" + stage + ")")
      if (!waitingStages(stage) && !runningStages(stage) && !failedStages(stage)) {
        val missing = getMissingParentStages(stage).sortBy(_.id) // 获取当前stage的父stage
        logDebug("missing: " + missing)
        // 没有获取到新的stage，结束stage划分，提交task
        if (missing.isEmpty) {
          logInfo("Submitting " + stage + " (" + stage.rdd + "), which has no missing parents")
          submitMissingTasks(stage, jobId.get)
        } else {
          for (parent <- missing) {
            // 获取到了新的父stage，递归调用自身，继续获取stage
            submitStage(parent)
          }
          waitingStages += stage // 将stage加入等待队列
        }
      }
    } else {
      abortStage(stage, "No active job for stage " + stage.id, None)
    }
  }

六 getMissingParentStages。根据是否是宽依赖，处理stage的划分。

对于每一种有shuffle的操作，比如groupByKey、reduceByKey、countByKey，底层对应了三个RDD: MapPartitionsRDD/ShuffleRDD/MapPartitionsRDD，前两个之间发生了shuffle，形成了宽依赖。

  private def getMissingParentStages(stage: Stage): List[Stage] = {
    val missing = new HashSet[Stage]
    val visited = new HashSet[RDD[_]]
    // We are manually maintaining a stack here to prevent StackOverflowError
    // caused by recursively visiting
    val waitingForVisit = new ArrayStack[RDD[_]]
    def visit(rdd: RDD[_]) {
      if (!visited(rdd)) {
        visited += rdd
        val rddHasUncachedPartitions = getCacheLocs(rdd).contains(Nil)
        if (rddHasUncachedPartitions) {
          for (dep <- rdd.dependencies) {
            // 遍历每一个rdd，只有宽依赖，才会创建stage
            dep match {
              case shufDep: ShuffleDependency[_, _, _] =>
                // 宽依赖，创建一个shuffleMapStage
                // 除了最后一个stage，其余的都是shuffleMapStage
                val mapStage = getOrCreateShuffleMapStage(shufDep, stage.firstJobId)
                if (!mapStage.isAvailable) {
                  missing += mapStage
                }
              case narrowDep: NarrowDependency[_] =>
                waitingForVisit.push(narrowDep.rdd) // 如果是窄依赖，将依赖的rdd放入栈中
            }
          }
        }
      }
    }
    waitingForVisit.push(stage.rdd) // 首先往栈中推入了finalStage rdd
    while (waitingForVisit.nonEmpty) {
      visit(waitingForVisit.pop())
    }
    missing.toList
  }

七 首先为第一个stage创建并提交task，然后依次对等待队列中的stage创建并提交task。

  /** Called when stage's parents are available and we can now do its task. */
  private def submitMissingTasks(stage: Stage, jobId: Int) {
    logDebug("submitMissingTasks(" + stage + ")")

    // First figure out the indexes of partition ids to compute.
    val partitionsToCompute: Seq[Int] = stage.findMissingPartitions()

    // Use the scheduling pool, job group, description, etc. from an ActiveJob associated
    // with this Stage
    val properties = jobIdToActiveJob(jobId).properties

    runningStages += stage
    // SparkListenerStageSubmitted should be posted before testing whether tasks are
    // serializable. If tasks are not serializable, a SparkListenerStageCompleted event
    // will be posted, which should always come after a corresponding SparkListenerStageSubmitted
    // event.
    stage match {
      case s: ShuffleMapStage =>
        outputCommitCoordinator.stageStart(stage = s.id, maxPartitionId = s.numPartitions - 1)
      case s: ResultStage =>
        outputCommitCoordinator.stageStart(
          stage = s.id, maxPartitionId = s.rdd.partitions.length - 1)
    }
    val taskIdToLocations: Map[Int, Seq[TaskLocation]] = try {
      stage match {
        case s: ShuffleMapStage =>
          partitionsToCompute.map { id => (id, getPreferredLocs(stage.rdd, id))}.toMap
        case s: ResultStage =>
          partitionsToCompute.map { id =>
            val p = s.partitions(id)
            (id, getPreferredLocs(stage.rdd, p))
          }.toMap
      }
    } catch {
      case NonFatal(e) =>
        stage.makeNewStageAttempt(partitionsToCompute.size)
        listenerBus.post(SparkListenerStageSubmitted(stage.latestInfo, properties))
        abortStage(stage, s"Task creation failed: $e\n${Utils.exceptionString(e)}", Some(e))
        runningStages -= stage
        return
    }

    stage.makeNewStageAttempt(partitionsToCompute.size, taskIdToLocations.values.toSeq)

    // If there are tasks to execute, record the submission time of the stage. Otherwise,
    // post the even without the submission time, which indicates that this stage was
    // skipped.
    if (partitionsToCompute.nonEmpty) {
      stage.latestInfo.submissionTime = Some(clock.getTimeMillis())
    }
    listenerBus.post(SparkListenerStageSubmitted(stage.latestInfo, properties))

    // TODO: Maybe we can keep the taskBinary in Stage to avoid serializing it multiple times.
    // Broadcasted binary for the task, used to dispatch tasks to executors. Note that we broadcast
    // the serialized copy of the RDD and for each task we will deserialize it, which means each
    // task gets a different copy of the RDD. This provides stronger isolation between tasks that
    // might modify state of objects referenced in their closures. This is necessary in Hadoop
    // where the JobConf/Configuration object is not thread-safe.
    var taskBinary: Broadcast[Array[Byte]] = null
    var partitions: Array[Partition] = null
    try {
      // For ShuffleMapTask, serialize and broadcast (rdd, shuffleDep).
      // For ResultTask, serialize and broadcast (rdd, func).
      var taskBinaryBytes: Array[Byte] = null
      // taskBinaryBytes and partitions are both effected by the checkpoint status. We need
      // this synchronization in case another concurrent job is checkpointing this RDD, so we get a
      // consistent view of both variables.
      RDDCheckpointData.synchronized {
        taskBinaryBytes = stage match {
          case stage: ShuffleMapStage =>
            JavaUtils.bufferToArray(
              closureSerializer.serialize((stage.rdd, stage.shuffleDep): AnyRef))
          case stage: ResultStage =>
            JavaUtils.bufferToArray(closureSerializer.serialize((stage.rdd, stage.func): AnyRef))
        }

        partitions = stage.rdd.partitions
      }

      taskBinary = sc.broadcast(taskBinaryBytes)
    } catch {
      // In the case of a failure during serialization, abort the stage.
      case e: NotSerializableException =>
        abortStage(stage, "Task not serializable: " + e.toString, Some(e))
        runningStages -= stage

        // Abort execution
        return
      case NonFatal(e) =>
        abortStage(stage, s"Task serialization failed: $e\n${Utils.exceptionString(e)}", Some(e))
        runningStages -= stage
        return
    }

    // 为stage创建指定数量的task
    val tasks: Seq[Task[_]] = try {
      val serializedTaskMetrics = closureSerializer.serialize(stage.latestInfo.taskMetrics).array()
      stage match {
        case stage: ShuffleMapStage =>
          stage.pendingPartitions.clear()
          partitionsToCompute.map { id =>
                  // 给每一个partition创建一个task
                  // 给每个task计算最佳位置
            val locs = taskIdToLocations(id)
            val part = partitions(id)
            stage.pendingPartitions += id
            // 非finalStage创建ShuffleMapTask
            new ShuffleMapTask(stage.id, stage.latestInfo.attemptNumber,
              taskBinary, part, locs, properties, serializedTaskMetrics, Option(jobId),
              Option(sc.applicationId), sc.applicationAttemptId)
          }

        case stage: ResultStage =>
          partitionsToCompute.map { id =>
            val p: Int = stage.partitions(id)
            val part = partitions(p)
            val locs = taskIdToLocations(id)
            // finalStage创建ResultTask
            new ResultTask(stage.id, stage.latestInfo.attemptNumber,
              taskBinary, part, locs, id, properties, serializedTaskMetrics,
              Option(jobId), Option(sc.applicationId), sc.applicationAttemptId)
          }
      }
    } catch {
      case NonFatal(e) =>
        abortStage(stage, s"Task creation failed: $e\n${Utils.exceptionString(e)}", Some(e))
        runningStages -= stage
        return
    }

    if (tasks.size > 0) {
      logInfo(s"Submitting ${tasks.size} missing tasks from $stage (${stage.rdd}) (first 15 " +
        s"tasks are for partitions ${tasks.take(15).map(_.partitionId)})")
      // 针对stage的task，创建TaskSet对象，调用TaskScheduler的submitTasks()方法，提交TaskSet
      taskScheduler.submitTasks(new TaskSet(
        tasks.toArray, stage.id, stage.latestInfo.attemptNumber, jobId, properties))
    } else {
      // Because we posted SparkListenerStageSubmitted earlier, we should mark
      // the stage as completed here in case there are no tasks to run
      markStageAsFinished(stage, None)

      val debugString = stage match {
        case stage: ShuffleMapStage =>
          s"Stage ${stage} is actually done; " +
            s"(available: ${stage.isAvailable}," +
            s"available outputs: ${stage.numAvailableOutputs}," +
            s"partitions: ${stage.numPartitions})"
        case stage : ResultStage =>
          s"Stage ${stage} is actually done; (partitions: ${stage.numPartitions})"
      }
      logDebug(debugString)

      submitWaitingChildStages(stage) // 调度等待队列
    }
  }

1 调度等待队列

  /**
   * Check for waiting stages which are now eligible for resubmission.
   * Submits stages that depend on the given parent stage. Called when the parent stage completes
   * successfully.
   */
  private def submitWaitingChildStages(parent: Stage) {
    logTrace(s"Checking if any dependencies of $parent are now runnable")
    logTrace("running: " + runningStages)
    logTrace("waiting: " + waitingStages)
    logTrace("failed: " + failedStages)
    val childStages = waitingStages.filter(_.parents.contains(parent)).toArray
    waitingStages --= childStages
    for (stage <- childStages.sortBy(_.firstJobId)) {
      submitStage(stage)
    }
  }

2 为task获取最佳位置

  /**
   * Gets the locality information associated with a partition of a particular RDD.
   *
   * This method is thread-safe and is called from both DAGScheduler and SparkContext.
   *
   * @param rdd whose partitions are to be looked at
   * @param partition to lookup locality information for
   * @return list of machines that are preferred by the partition
   */
  private[spark]
  def getPreferredLocs(rdd: RDD[_], partition: Int): Seq[TaskLocation] = {
    getPreferredLocsInternal(rdd, partition, new HashSet)
  }

  /**
   * Recursive implementation for getPreferredLocs.
   *
   * This method is thread-safe because it only accesses DAGScheduler state through thread-safe
   * methods (getCacheLocs()); please be careful when modifying this method, because any new
   * DAGScheduler state accessed by it may require additional synchronization.
   */
  // 优先使用缓存或checkpoint的节点，因为这个节点之前的rdd不需要重新计算了
  private def getPreferredLocsInternal(
      rdd: RDD[_],
      partition: Int,
      visited: HashSet[(RDD[_], Int)]): Seq[TaskLocation] = {
    // If the partition has already been visited, no need to re-visit.
    // This avoids exponential path exploration.  SPARK-695
    if (!visited.add((rdd, partition))) {
      // Nil has already been returned for previously visited partitions.
      return Nil
    }
    // If the partition is cached, return the cache locations
    val cached = getCacheLocs(rdd)(partition)
    if (cached.nonEmpty) {
      return cached
    }
    // If the RDD has some placement preferences (as is the case for input RDDs), get those
    val rddPrefs = rdd.preferredLocations(rdd.partitions(partition)).toList
    if (rddPrefs.nonEmpty) {
      return rddPrefs.map(TaskLocation(_))
    }

    // If the RDD has narrow dependencies, pick the first partition of the first narrow dependency
    // that has any placement preferences. Ideally we would choose based on transfer sizes,
    // but this will do for now.
    rdd.dependencies.foreach {
      case n: NarrowDependency[_] =>
        for (inPart <- n.getParents(partition)) {
          val locs = getPreferredLocsInternal(n.rdd, inPart, visited)
          if (locs != Nil) {
            return locs
          }
        }

      case _ =>
    }

    // 否则没有最佳位置
    Nil
  }