本帖最后由 pig2 于 2015-1-6 14:17 编辑
问题导读
1.构成Standalone cluster部署模式的四大组成部件有哪些?分别有什么功能?
2.WorkerInfo在schedule函数中会被使用到,schedule函数处理逻辑是怎样的?
概要本文主要讲述在standalone cluster部署模式下,Spark Application在整个运行期间,资源(主要是cpu core和内存)的申请与释放。 构成Standalone cluster部署模式的四大组成部件如下图所示,分别为Master, worker, executor和driver,它们各自运行于独立的JVM进程。
从资源管理的角度来说 - Master 掌管整个cluster的资源,主要是指cpu core和memory,但Master自身并不拥有这些资源
- Worker 计算资源的实际贡献者,须向Master汇报自身拥有多少cpu core和memory, 在master的指示下负责启动executor
- Executor 执行真正计算的苦力,由master来决定该进程拥有的core和memory数值
- Driver 资源的实际占用者,Driver会提交一到多个job,每个job在拆分成多个task之后,会分发到各个executor真正的执行
这些内容在standalone cluster模式下的容错性分析中也有所涉及,今天主要讲一下资源在分配之后不同场景下是如何被顺利回收的。
资源上报汇聚过程standalone cluster下最主要的当然是master,master必须先于worker和driver程序正常启动。 当master顺利启动完毕,可以开始worker的启动工作,worker在启动的时候需要向master发起注册,在注册消息中带有本worker节点的cpu core和内存。 调用顺序如下preStart->registerWithMaster->tryRegisterAllMasters 看一看tryRegisterAllMasters的代码
- def tryRegisterAllMasters() {
- for (masterUrl <- masterUrls) {
- logInfo("Connecting to master " + masterUrl + "...")
- val actor = context.actorSelection(Master.toAkkaUrl(masterUrl))
- actor ! RegisterWorker(workerId, host, port, cores, memory, webUi.boundPort, publicAddress)
- }
- }
复制代码
我们的疑问是RegisterWorker构造函数所需的参数memory和cores是从哪里获取的呢? 注意一下Worker中的main函数会创建WorkerArguments,
- def main(argStrings: Array[String]) {
- SignalLogger.register(log)
- val args = new WorkerArguments(argStrings)
- val (actorSystem, _) = startSystemAndActor(args.host, args.port, args.webUiPort, args.cores,
- args.memory, args.masters, args.workDir)
- actorSystem.awaitTermination()
- }
复制代码
memory通过函数inferDefaultMemory获取,而cores通过inferDefaultCores获取。
- def inferDefaultCores(): Int = {
- Runtime.getRuntime.availableProcessors()
- }
-
- def inferDefaultMemory(): Int = {
- val ibmVendor = System.getProperty("java.vendor").contains("IBM")
- var totalMb = 0
- try {
- val bean = ManagementFactory.getOperatingSystemMXBean()
- if (ibmVendor) {
- val beanClass = Class.forName("com.ibm.lang.management.OperatingSystemMXBean")
- val method = beanClass.getDeclaredMethod("getTotalPhysicalMemory")
- totalMb = (method.invoke(bean).asInstanceOf[Long] / 1024 / 1024).toInt
- } else {
- val beanClass = Class.forName("com.sun.management.OperatingSystemMXBean")
- val method = beanClass.getDeclaredMethod("getTotalPhysicalMemorySize")
- totalMb = (method.invoke(bean).asInstanceOf[Long] / 1024 / 1024).toInt
- }
- } catch {
- case e: Exception => {
- totalMb = 2*1024
- System.out.println("Failed to get total physical memory. Using " + totalMb + " MB")
- }
- }
- // Leave out 1 GB for the operating system, but don't return a negative memory size
- math.max(totalMb - 1024, 512)
- }
复制代码
如果已经在配置文件中为显示指定了每个worker的core和memory,则使用配置文件中的值,具体配置参数为SPARK_WORKER_CORES和SPARK_WORKER_MEMORY
Master在收到RegisterWork消息之后,根据上报的信息为每一个worker创建相应的WorkerInfo.
- case RegisterWorker(id, workerHost, workerPort, cores, memory, workerUiPort, publicAddress) =>
- {
- logInfo("Registering worker %s:%d with %d cores, %s RAM".format(
- workerHost, workerPort, cores, Utils.megabytesToString(memory)))
- if (state == RecoveryState.STANDBY) {
- // ignore, don't send response
- } else if (idToWorker.contains(id)) {
- sender ! RegisterWorkerFailed("Duplicate worker ID")
- } else {
- val worker = new WorkerInfo(id, workerHost, workerPort, cores, memory,
- sender, workerUiPort, publicAddress)
- if (registerWorker(worker)) {
- persistenceEngine.addWorker(worker)
- sender ! RegisteredWorker(masterUrl, masterWebUiUrl)
- schedule()
- } else {
- val workerAddress = worker.actor.path.address
- logWarning("Worker registration failed. Attempted to re-register worker at same " +
- "address: " + workerAddress)
- sender ! RegisterWorkerFailed("Attempted to re-register worker at same address: "
- + workerAddress)
- }
- }
复制代码
资源分配过程如果在worker注册上来的时候,已经有Driver Application注册上来,那么就需要将原先处于未分配资源状态的driver application启动相应的executor。 WorkerInfo在schedule函数中会被使用到,schedule函数处理逻辑概述如下 - 查看目前存活的worker中剩余的内存是否能够满足application每个task的最低需求,如果是则将该worker加入到可分配资源的队列
- 根据分发策略,如果是决定将工作平摊到每个worker,则每次在一个worker上占用一个core,直到所有可分配资源耗尽或已经满足driver的需求
- 如果分发策略是分发到尽可能少的worker,则一次占用尽worker上的可分配core,直到driver的core需求得到满足
- 根据步骤2或3的结果在每个worker上添加相应的executor,处理函数是addExecutor
为了叙述简单,现仅列出平摊到各个worker的分配处理过程
- for (worker > workers if worker.coresFree > 0 && worker.state == WorkerState.ALIVE) {
- for (app <- waitingApps if app.coresLeft > 0) {
- if (canUse(app, worker)) {
- val coresToUse = math.min(worker.coresFree, app.coresLeft)
- if (coresToUse > 0) {
- val exec = app.addExecutor(worker, coresToUse)
- launchExecutor(worker, exec)
- app.state = ApplicationState.RUNNING
- }
- }
- }
- }
复制代码
launchExecutor主要负责两件事情 - 记录下新添加的executor使用掉的cpu core和内存数目,记录过程发生在worker.addExecutor
- 向worker发送LaunchExecutor指令
- def launchExecutor(worker: WorkerInfo, exec: ExecutorInfo) {
- logInfo("Launching executor " + exec.fullId + " on worker " + worker.id)
- worker.addExecutor(exec)
- worker.actor ! LaunchExecutor(masterUrl,
- exec.application.id, exec.id, exec.application.desc, exec.cores, exec.memory)
- exec.application.driver ! ExecutorAdded(
- exec.id, worker.id, worker.hostPort, exec.cores, exec.memory)
- }
复制代码
worker在收到LaunchExecutor指令后,也会记一笔账,将要使用掉的cpu core和memory从可用资源中减去,然后使用ExecutorRunner来负责生成Executor进程,注意Executor运行于独立的进程。代码如下
- case LaunchExecutor(masterUrl, appId, execId, appDesc, cores_, memory_) =>
- if (masterUrl != activeMasterUrl) {
- logWarning("Invalid Master (" + masterUrl + ") attempted to launch executor.")
- } else {
- try {
- logInfo("Asked to launch executor %s/%d for %s".format(appId, execId, appDesc.name))
- val manager = new ExecutorRunner(appId, execId, appDesc, cores_, memory_,
- self, workerId, host,
- appDesc.sparkHome.map(userSparkHome => new File(userSparkHome)).getOrElse(sparkHome),
- workDir, akkaUrl, conf, ExecutorState.RUNNING)
- executors(appId + "/" + execId) = manager
- manager.start()
- coresUsed += cores_
- memoryUsed += memory_
- masterLock.synchronized {
- master ! ExecutorStateChanged(appId, execId, manager.state, None, None)
- }
- } catch {
- case e: Exception => {
- logError("Failed to launch executor %s/%d for %s".format(appId, execId, appDesc.name))
- if (executors.contains(appId + "/" + execId)) {
- executors(appId + "/" + execId).kill()
- executors -= appId + "/" + execId
- }
- masterLock.synchronized {
- master ! ExecutorStateChanged(appId, execId, ExecutorState.FAILED, None, None)
- }
- }
- }
- }
复制代码
在资源分配过程中需要注意到的是如果有多个Driver Application处于等待状态,资源分配的原则是FIFO,先到先得。
资源回收过程worker中上报的资源最终被driver application中提交的job task所占用,如果application结束(包括正常和异常退出),application所占用的资源就应该被顺利回收,即将占用的资源重新归入可分配资源行列。 现在的问题转换成Master和Executor如何知道Driver Application已经退出了呢? 有两种不同的处理方式,一种是先道别后离开,一种是不告而别。现分别阐述。 何为先道别后离开,即driver application显式的通知master和executor,任务已经完成了,我要bye了。应用程序显式的调用SparkContext.stop
- def stop() {
- postApplicationEnd()
- ui.stop()
- // Do this only if not stopped already - best case effort.
- // prevent NPE if stopped more than once.
- val dagSchedulerCopy = dagScheduler
- dagScheduler = null
- if (dagSchedulerCopy != null) {
- metadataCleaner.cancel()
- cleaner.foreach(_.stop())
- dagSchedulerCopy.stop()
- taskScheduler = null
- // TODO: Cache.stop()?
- env.stop()
- SparkEnv.set(null)
- ShuffleMapTask.clearCache()
- ResultTask.clearCache()
- listenerBus.stop()
- eventLogger.foreach(_.stop())
- logInfo("Successfully stopped SparkContext")
- } else {
- logInfo("SparkContext already stopped")
- }
- }
复制代码
显式调用SparkContext.stop的一个主要功能是会去显式的停止Executor,具体下达StopExecutor指令的代码见于CoarseGrainedSchedulerBackend中的stop函数
- override def stop() {
- stopExecutors()
- try {
- if (driverActor != null) {
- val future = driverActor.ask(StopDriver)(timeout)
- Await.ready(future, timeout)
- }
- } catch {
- case e: Exception =>
- throw new SparkException("Error stopping standalone scheduler's driver actor", e)
- }
- }
复制代码
那么Master又是如何知道Driver Application退出的呢?这要归功于Akka的通讯机制了,当相互通讯的任意一方异常退出,另一方都会收到DisassociatedEvent, Master也就是在这个消息处理中移除已经停止的Driver Application。
- case DisassociatedEvent(_, address, _) => {
- // The disconnected client could've been either a worker or an app; remove whichever it was
- logInfo(s"$address got disassociated, removing it.")
- addressToWorker.get(address).foreach(removeWorker)
- addressToApp.get(address).foreach(finishApplication)
- if (state == RecoveryState.RECOVERING && canCompleteRecovery) { completeRecovery() }
- }
复制代码
不告而别的方式下Executor是如何知道自己所服务的application已经顺利完成使命了呢?道理和master的一样,还是通过DisassociatedEvent来感知。详见CoarseGrainedExecutorBackend中的receive函数
- case x: DisassociatedEvent =>
- logError(s"Driver $x disassociated! Shutting down.")
- System.exit(1)
复制代码
异常情况下的资源回收由于Master和Worker之间的心跳机制,如果worker异常退出, Master会由心跳机制感知到其消亡,进而将其上报的资源移除。 Executor异常退出时,Worker中的监控线程ExecutorRunner会立即感知,进而上报给Master,Master会回收资源,并重新要求worker启动executor。
相关内容
Apache Spark源码走读之1 -- Spark论文阅读笔记
Apache Spark源码走读之2 -- Job的提交与运行
Apache Spark源码走读之3-- Task运行期之函数调用关系分析
Apache Spark源码走读之4 -- DStream实时流数据处理
Apache Spark源码走读之5-- DStream处理的容错性分析
Apache Spark源码走读之6-- 存储子系统分析
Apache Spark源码走读之7 -- Standalone部署方式分析
Apache Spark源码走读之8 -- Spark on Yarn
Apache Spark源码走读之9 -- Spark源码编译
Apache Spark源码走读之10 -- 在YARN上运行SparkPi
Apache Spark源码走读之11 -- sql的解析与执行
Apache Spark源码走读之12 -- Hive on Spark运行环境搭建
Apache Spark源码走读之13 -- hiveql on spark实现详解
Apache Spark源码走读之14 -- Graphx实现剖析
Apache Spark源码走读之15 -- Standalone部署模式下的容错性分析
Apache Spark源码走读之16 -- spark repl实现详解
Apache Spark源码走读之17 -- 如何进行代码跟读
Apache Spark源码走读之18 -- 使用Intellij idea调试Spark源码
Apache Spark源码走读之20 -- ShuffleMapTask计算结果的保存与读取
Apache Spark源码走读之21 -- WEB UI和Metrics初始化及数据更新过程分析
Apache Spark源码走读之22 -- 浅谈mllib中线性回归的算法实现
Apache Spark源码走读之23 -- Spark MLLib中拟牛顿法L-BFGS的源码实现
Apache Spark源码走读之24 -- Sort-based Shuffle的设计与实现
|