spark中spark-submit内部实现原理.doc

这两天在看spark-submit之后任务是怎么启动的，写篇文章记录一下自己的理解思路： SparkSubmit(SparkDeploySchedulerBackend(AppClient(tryRegisterAllMaster 1:客户端启动，初始化相关的环境变量，包括Application代码的提交 2:向Master注册Driver 这里需要注意，Master，Worker是已经启动，换句话说，我们的spark集群已经启动。 3: SparkDeploySchedulerBackend中启动Application。向Master注册Application def registerWithMaster() { tryRegisterAllMasters() import context.dispatcher var retries = 0 registrationRetryTimer = Some { context.system.scheduler.schedule(REGISTRATION_TIMEOUT, REGISTRATION_TIMEOUT) { Utils.tryOrExit { retries += 1 if (registered) { registrationRetryTimer.foreach(_.cancel()) } else if (retries = REGISTRATION_RETRIES) { markDead(All masters are unresponsive! Giving up.) } else { tryRegisterAllMasters() } } } } } def tryRegisterAllMasters() { for (masterAkkaUrl - masterAkkaUrls) { logInfo(Connecting to master + masterAkkaUrl + ...) val actor = context.actorSelection(masterAkkaUrl) actor ! RegisterApplication(appDescription) } } 主要是这两段代码：SparkDeploySchedulerBackend中有Application的start函数，进入到APPClient中之后，首先向master注册Application，就是上面代码中的tryRegisterAllMasters()，这里会向Master发送RegisterApplication(appDescription)。在Master端通过AKKA收到消息后处理该消息，处理代码如下： case RegisterApplication(description) = { if (state == RecoveryState.STANDBY) { // ignore, dont send response } else { logInfo(Registering app + ) val app = createApplication(description, sender) registerApplication(app) logInfo(Registered app + + with ID + app.id) persistenceEngine.addApplication(app) sender ! RegisteredApplication(app.id, masterUrl) schedule() } } 进行注册Application，并且发送已经注册App的消息。然后进行schedule()。 Schedule函数如下： /** * Schedule the currently available resources among waiting apps. This method will be called * every t