我的Spark源码核心SparkContext走读全纪录

Posted 2020-08-09

我的Spark源码核心SparkContext走读全纪录

Dirver Program(SparkConf)  package org.apache.spark.SparkConf

Master        package org.apache.spark.deploy.master

SparkContext  package org.apache.spark.SparkContext

Stage         package org.apache.spark.scheduler.Stage

Task          package org.apache.spark.scheduler.Task  

DAGScheduler  package org.apache.spark.scheduler   

TaskScheduler package org.apache.spark.scheduler.TaskScheduler

TaskSchedulerImpl  package org.apache.spark.scheduler

Worker        package org.apache.spark.deploy.worker

Executor      package org.apache.spark.executor

BlockManager  package org.apache.spark.storage

TaskSet       package org.apache.spark.scheduler

//初始化后开始创建

// Create and start the scheduler

    val (sched, ts) = SparkContext.createTaskScheduler(this, master)

    _schedulerBackend = sched

    _taskScheduler = ts

    _dagScheduler = new DAGScheduler(this)

    _heartbeatReceiver.send(TaskSchedulerIsSet)

 

/**

   * Create a task scheduler based on a given master URL.

   * Return a 2-tuple of the scheduler backend and the task scheduler.

   */

  private def createTaskScheduler(

      sc: SparkContext,

      master: String): (SchedulerBackend, TaskScheduler) = {

master match {

      case "local" =>

实例化一个

val scheduler = new TaskSchedulerImpl(sc)

构建masterUrls：

val masterUrls = localCluster.start()

据说是非常关键的backend：

val backend = new SparkDeploySchedulerBackend(scheduler, sc, masterUrls)

        scheduler.initialize(backend)

        backend.shutdownCallback = (backend: SparkDeploySchedulerBackend) => {

          localCluster.stop()

        }

        (backend, scheduler)