DAGScheduler
DAGScheduler的主要任务是基于Stage构建DAG，决定每个任务的最佳位置
- 记录哪个RDD或者Stage输出被物化
- 面向stage的调度层，为job生成以stage组成的DAG，提交TaskSet给TaskScheduler执行
- 重新提交shuffle输出丢失的stage
每一个Stage内，都是独立的tasks，他们共同执行同一个computefunction，享有相同的shuffledependencies。DAG在切分stage的时候是依照出现shuffle为界限的。
DAGScheduler实例化
下面的代码是SparkContext实例化DAGScheduler的过程：
@volatile private[spark] var dagScheduler: DAGScheduler = _
try {
dagScheduler = new DAGScheduler(this)
} catch {
case e: Exception => {
try {
stop()
} finally {
throw new SparkException("Error while constructing DAGScheduler", e)
}
}
}
下面代码显示了DAGScheduler的构造函数定义中，通过绑定TaskScheduler的方式创建，其中次构造函数去调用主构造函数来将sc的字段填充入参：
private[spark]
class DAGScheduler(
private[scheduler] val sc: SparkContext,
private[scheduler] val taskScheduler: TaskScheduler,
listenerBus: LiveListenerBus,
mapOutputTracker: MapOutputTrackerMaster,
blockManagerMaster: BlockManagerMaster,
env: SparkEnv,
clock: Clock = new SystemClock())
extends Logging {
def this(sc: SparkContext, taskScheduler: TaskScheduler) = {
this(
sc,
taskScheduler,
sc.listenerBus,
sc.env.mapOutputTracker.asInstanceOf[MapOutputTrackerMaster],
sc.env.blockManager.master,
sc.env)
}
def this(sc: SparkContext) = this(sc, sc.taskScheduler)
作业提交与DAGScheduler操作
Action的大部分操作会进行作业（job）的提交，源码1.0版的job提交过程的大致调用链是：sc.runJob()–>dagScheduler.runJob–>dagScheduler.submitJob—>dagSchedulerEventProcessActor.JobSubmitted–>dagScheduler.handleJobSubmitted–>dagScheduler.submitStage–>dagScheduler.submitMissingTasks–>taskScheduler.submitTasks。
具体的作业提交执行期的函数调用为：
- sc.runJob->dagScheduler.runJob->submitJob
- DAGScheduler::submitJob会创建JobSummitted的event发送给内嵌类eventProcessActor（在源码1.4中，submitJob函数中，使用DAGSchedulerEventProcessLoop类进行事件的处理）
- eventProcessActor在接收到JobSubmmitted之后调用processEvent处理函数
- job到stage的转换，生成finalStage并提交运行，关键是调用submitStage
- 在submitStage中会计算stage之间的依赖关系，依赖关系分为宽依赖和窄依赖两种
- 如果计算中发现当前的stage没有任何依赖或者所有的依赖都已经准备完毕，则提交task
- 提交task是调用函数submitMissingTasks来完成
- task真正运行在哪个worker上面是由TaskScheduler来管理，也就是上面的submitMissingTasks会调用TaskScheduler::submitTasks
- TaskSchedulerImpl中会根据Spark的当前运行模式来创建相应的backend,如果是在单机运行则创建LocalBackend
- LocalBackend收到TaskSchedulerImpl传递进来的ReceiveOffers事件
- receiveOffers->executor.launchTask->TaskRunner.run
DAGScheduler的runJob函数
DAGScheduler.runjob最后把结果通过resultHandler保存返回。
这里DAGScheduler的runJob函数调用DAGScheduler的submitJob函数来提交任务：
def runJob[T, U: ClassTag](
rdd: RDD[T],
func: (TaskContext, Iterator[T]) => U,
partitions: Seq[Int],
callSite: CallSite,
allowLocal: Boolean,
resultHandler: (Int, U) => Unit,
properties: Properties): Unit = {
val start = System.nanoTime
val waiter = submitJob(rdd, func, partitions, callSite, allowLocal, resultHandler, properties)
waiter.awaitResult() match {
case JobSucceeded => {
logInfo("Job %d finished: %s, took %f s".format
(waiter.jobId, callSite.shortForm, (System.nanoTime - start) / 1e9))
}
case JobFailed(exception: Exception) =>
logInfo("Job %d failed: %s, took %f s".format
(waiter.jobId, callSite.shortForm, (System.nanoTime - start) / 1e9))
throw exception
}
}
作业提交的调度
在Spark源码1.4.0中，DAGScheduler的submitJob函数不再使用DAGEventProcessActor进行事件处理和消息通信，而是使用DAGSchedulerEventProcessLoop类实例eventProcessLoop进行JobSubmitted事件的post动作。
下面是submitJob函数代码：
/**
* Submit a job to the job scheduler and get a JobWaiter object back. The JobWaiter object
* can be used to block until the the job finishes executing or can be used to cancel the job.
*/
def submitJob[T, U](
rdd: RDD[T],
func: (TaskContext, Iterator[T]) => U,
partitions: Seq[Int],
callSite: CallSite,
allowLocal: Boolean,
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 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)
eventProcessLoop.post(JobSubmitted(
jobId, rdd, func2, partitions.toArray, allowLocal, callSite, waiter, properties))
waiter
}
当eventProcessLoop对象投递了JobSubmitted事件之后，对象内的eventThread线程实例对事件进行处理，不断从事件队列中取出事件，调用onReceive函数处理事件，当匹配到JobSubmitted事件后，调用DAGScheduler的handleJobSubmitted函数并传入jobid、rdd等参数来处理Job。
handleJobSubmitted函数
Job处理过程中handleJobSubmitted比较关键，该函数主要负责RDD的依赖性分析，生成finalStage，并根据finalStage来产生ActiveJob。
在handleJobSubmitted函数源码中，给出了部分注释：
private[scheduler] def handleJobSubmitted(jobId: Int,
finalRDD: RDD[_],
func: (TaskContext, Iterator[_]) => _,
partitions: Array[Int],
allowLocal: Boolean,
callSite: CallSite,
listener: JobListener,
properties: Properties) {
var finalStage: Stage = 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 = newStage(finalRDD, partitions.size, None, jobId, callSite)
} catch {
//错误处理，告诉监听器作业失败，返回....
case e: Exception =>
logWarning("Creating new stage failed due to exception - job: " + jobId, e)
listener.jobFailed(e)
return
}
if (finalStage != null) {
val job = new ActiveJob(jobId, finalStage, func, partitions, callSite, listener, properties)
clearCacheLocs()
logInfo("Got job %s (%s) with %d output partitions (allowLocal=%s)".format(
job.jobId, callSite.shortForm, partitions.length, allowLocal))
logInfo("Final stage: " + finalStage + "(" + finalStage.name + ")")
logInfo("Parents of final stage: " + finalStage.parents)
logInfo("Missing parents: " + getMissingParentStages(finalStage))
val shouldRunLocally =
localExecutionEnabled && allowLocal && finalStage.parents.isEmpty && partitions.length == 1
val jobSubmissionTime = clock.getTimeMillis()
if (shouldRunLocally) {
// 很短、没有父stage的本地操作，比如 first() or take() 的操作本地执行
// Compute very short actions like first() or take() with no parent stages locally.
listenerBus.post(
SparkListenerJobStart(job.jobId, jobSubmissionTime, Seq.empty, properties))
runLocally(job)
} else {
// collect等操作走的是这个过程，更新相关的关系映射，用监听器监听，然后提交作业
jobIdToActiveJob(jobId) = job
activeJobs += job
finalStage.resultOfJob = Some(job)
val stageIds = jobIdToStageIds(jobId).toArray
val stageInfos = stageIds.flatMap(id => stageIdToStage.get(id).map(_.latestInfo))
listenerBus.post(
SparkListenerJobStart(job.jobId, jobSubmissionTime, stageInfos, properties))
// 提交stage
submitStage(finalStage)
}
}
// 提交stage
submitWaitingStages()
}
小结
该篇文章介绍了DAGScheduler从SparkContext中进行实例化，到执行Action操作时提交任务调用runJob函数，进而介绍了提交任务的消息调度，和处理Job函数handleJobSubmitted函数。
由于在handleJobSubmitted函数中涉及到依赖性分析和stage的源码内容，于是我计划在下一篇文章里进行介绍和源码分析。
转载请注明作者Jason Ding及其出处
GitCafe博客主页(http://jasonding1354.gitcafe.io/)
Github博客主页(http://jasonding1354.github.io/)
CSDN博客(http://blog.csdn.net/jasonding1354)
简书主页(http://www.jianshu.com/users/2bd9b48f6ea8/latest_articles)
Google搜索jasonding1354进入我的博客主页
