Spark源码分析之-deploy模块

architecture (6)

cloud 7

spark 8

30 April 2013

Background

在前文Spark源码分析之-scheduler模块中提到了Spark在资源管理和调度上采用了Hadoop YARN的方式：外层的资源管理器和应用内的任务调度器；并且分析了Spark应用内的任务调度模块。本文就Spark的外层资源管理器-deploy模块进行分析，探究Spark是如何协调应用之间的资源调度和管理的。
Spark最初是交由Mesos进行资源管理，为了使得更多的用户，包括没有接触过Mesos的用户使用Spark，Spark的开发者添加了Standalone的部署方式，也就是deploy模块。因此deploy模块只针对不使用Mesos进行资源管理的部署方式。

Deploy模块整体架构

deploy模块主要包含3个子模块：master, worker, client。他们继承于

Actor

，通过actor实现互相之间的通信。
Master：master的主要功能是接收worker的注册并管理所有的worker，接收client提交的application，(FIFO)调度等待的application并向worker提交。
Worker：worker的主要功能是向master注册自己，根据master发送的application配置进程环境，并启动

StandaloneExecutorBackend

。
Client：client的主要功能是向master注册并监控application。当用户创建

SparkContext

时会实例化

SparkDeploySchedulerBackend

，而实例化

SparkDeploySchedulerBackend

的同时就会启动client，通过向client传递启动参数和application有关信息，client向master发送请求注册application并且在slave
node上启动

StandaloneExecutorBackend

。
下面来看一下deploy模块的类图：

Deploy模块通信消息

Deploy模块并不复杂，代码也不多，主要集中在各个子模块之间的消息传递和处理上，因此在这里列出了各个模块之间传递的主要消息：

client to master

RegisterApplication

 (向master注册application)

master to client

RegisteredApplication

 (作为注册application的reply，回复给client)

ExecutorAdded

 (通知client worker已经启动了Executor环境，当向worker发送

LaunchExecutor

后通知client)

ExecutorUpdated

 (通知client Executor状态已经发生变化了，包括结束、异常退出等，当worker向master发送

ExecutorStateChanged

后通知client)

master to worker

LaunchExecutor

 (发送消息启动Executor环境)

RegisteredWorker

 (作为worker向master注册的reply)

RegisterWorkerFailed

 (作为worker向master注册失败的reply)

KillExecutor

 (发送给worker请求停止executor环境)

worker to master

RegisterWorker

 (向master注册自己)

Heartbeat

 (定期向master发送心跳信息)

ExecutorStateChanged

 (向master发送Executor状态改变信息)

Deploy模块代码详解

Deploy模块相比于scheduler模块简单，因此对于deploy模块的代码并不做十分细节的分析，只针对application的提交和结束过程做一定的分析。

Client提交application

Client是由

SparkDeploySchedulerBackend

创建被启动的，因此client是被嵌入在每一个application中，只为这个applicator所服务，在client启动时首先会先master注册application：

def start() {

// Just launch an actor; it will call back into the listener.

actor = actorSystem.actorOf(Props(new ClientActor))

}

override def preStart() {

logInfo("Connecting to master " + masterUrl)

try {

master = context.actorFor(Master.toAkkaUrl(masterUrl))

masterAddress = master.path.address

master ! RegisterApplication(appDescription) //向master注册application

context.system.eventStream.subscribe(self, classOf[RemoteClientLifeCycleEvent])

context.watch(master)  // Doesn't work with remote actors, but useful for testing

} catch {

case e: Exception =>

logError("Failed to connect to master", e)

markDisconnected()

context.stop(self)

}

}

Master在收到

RegisterApplication

请求后会把application加到等待队列中，等待调度：

case RegisterApplication(description) => {

logInfo("Registering app " + description.name)

val app = addApplication(description, sender)

logInfo("Registered app " + description.name + " with ID " + app.id)

waitingApps += app

context.watch(sender)  // This doesn't work with remote actors but helps for testing

sender ! RegisteredApplication(app.id)

schedule()

}

Master会在每次操作后调用

schedule()

函数，以确保等待的application能够被及时调度。
在前面提到deploy模块是资源管理模块，那么Spark的deploy管理的是什么资源，资源以什么单位进行调度的呢？在当前版本的Spark中，集群的cpu数量是Spark资源管理的一个标准，每个提交的application都会标明自己所需要的资源数(也就是cpu的core数)，Master以FIFO的方式管理所有的application请求，当资源数量满足当前任务执行需求的时候该任务就会被调度，否则就继续等待，当然如果master能给予当前任务部分资源则也会启动该application。

schedule()

函数实现的就是此功能。

def schedule() {

if (spreadOutApps) {

for (app <- waitingApps if app.coresLeft > 0) {

val usableWorkers = workers.toArray.filter(_.state == WorkerState.ALIVE)

.filter(canUse(app, _)).sortBy(_.coresFree).reverse

val numUsable = usableWorkers.length

val assigned = new Array[Int](numUsable) // Number of cores to give on each node

var toAssign = math.min(app.coresLeft, usableWorkers.map(_.coresFree).sum)

var pos = 0

while (toAssign > 0) {

if (usableWorkers(pos).coresFree - assigned(pos) > 0) {

toAssign -= 1

assigned(pos) += 1

}

pos = (pos + 1) % numUsable

}

// Now that we've decided how many cores to give on each node, let's actually give them

for (pos <- 0 until numUsable) {

if (assigned(pos) > 0) {

val exec = app.addExecutor(usableWorkers(pos), assigned(pos))

launchExecutor(usableWorkers(pos), exec, app.desc.sparkHome)

app.state = ApplicationState.RUNNING

}

}

}

} else {

// Pack each app into as few nodes as possible until we've assigned all its cores

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.desc.sparkHome)

  app.state = ApplicationState.RUNNING

}

}

}

}

}

}

当application得到调度后就会调用

launchExecutor()

向worker发送请求，同时向client汇报状态：

def launchExecutor(worker: WorkerInfo, exec: ExecutorInfo, sparkHome: String) {

worker.addExecutor(exec)

worker.actor ! LaunchExecutor(exec.application.id, exec.id, exec.application.desc, exec.cores, exec.memory, sparkHome)

exec.application.driver ! ExecutorAdded(exec.id, worker.id, worker.host, exec.cores, exec.memory)

}

至此client与master的交互已经转向了master与worker的交互，worker需要配置application启动环境

case LaunchExecutor(appId, execId, appDesc, cores_, memory_, execSparkHome_) =>

val manager = new ExecutorRunner(

appId, execId, appDesc, cores_, memory_, self, workerId, ip, new File(execSparkHome_), workDir)

executors(appId + "/" + execId) = manager

manager.start()

coresUsed += cores_

memoryUsed += memory_

master ! ExecutorStateChanged(appId, execId, ExecutorState.RUNNING, None, None)

Worker在接收到

LaunchExecutor

消息后创建

ExecutorRunner

实例，同时汇报master
executor环境启动。

ExecutorRunner

在启动的过程中会创建线程，配置环境，启动新进程：

def start() {

workerThread = new Thread("ExecutorRunner for " + fullId) {

override def run() { fetchAndRunExecutor() }

}

workerThread.start()

// Shutdown hook that kills actors on shutdown.

...

}

def fetchAndRunExecutor() {

try {

// Create the executor's working directory

val executorDir = new File(workDir, appId + "/" + execId)

if (!executorDir.mkdirs()) {

throw new IOException("Failed to create directory " + executorDir)

}

// Launch the process

val command = buildCommandSeq()

val builder = new ProcessBuilder(command: _*).directory(executorDir)

val env = builder.environment()

for ((key, value) <- appDesc.command.environment) {

env.put(key, value)

}

env.put("SPARK_MEM", memory.toString + "m")

// In case we are running this from within the Spark Shell, avoid creating a "scala"

// parent process for the executor command

env.put("SPARK_LAUNCH_WITH_SCALA", "0")

process = builder.start()

// Redirect its stdout and stderr to files

redirectStream(process.getInputStream, new File(executorDir, "stdout"))

redirectStream(process.getErrorStream, new File(executorDir, "stderr"))

// Wait for it to exit; this is actually a bad thing if it happens, because we expect to run

// long-lived processes only. However, in the future, we might restart the executor a few

// times on the same machine.

val exitCode = process.waitFor()

val message = "Command exited with code " + exitCode

worker ! ExecutorStateChanged(appId, execId, ExecutorState.FAILED, Some(message),

Some(exitCode))

} catch {

case interrupted: InterruptedException =>

logInfo("Runner thread for executor " + fullId + " interrupted")

case e: Exception => {

logError("Error running executor", e)

if (process != null) {

process.destroy()

}

val message = e.getClass + ": " + e.getMessage

  worker ! ExecutorStateChanged(appId, execId, ExecutorState.FAILED, Some(message), None)

}

}

}

在

ExecutorRunner

启动后worker向master汇报

ExecutorStateChanged

，而master则将消息重新pack成为

ExecutorUpdated

发送给client。
至此整个application提交过程基本结束，提交的过程并不复杂，主要涉及到的消息的传递。

Application的结束

由于各种原因(包括正常结束，异常返回等)会造成application的结束，我们现在就来看看applicatoin结束的整个流程。
application的结束往往会造成client的结束，而client的结束会被master通过

Actor

检测到，master检测到后会调用

removeApplication()

函数进行操作：

def removeApplication(app: ApplicationInfo) {

if (apps.contains(app)) {

logInfo("Removing app " + app.id)

apps -= app

idToApp -= app.id

actorToApp -= app.driver

addressToWorker -= app.driver.path.address

completedApps += app   // Remember it in our history

waitingApps -= app

for (exec <- app.executors.values) {

exec.worker.removeExecutor(exec)

exec.worker.actor ! KillExecutor(exec.application.id, exec.id)

}

app.markFinished(ApplicationState.FINISHED)  // TODO: Mark it as FAILED if it failed

  schedule()

}

}

removeApplicatoin()

首先会将application从master自身所管理的数据结构中删除，其次它会通知每一个work，请求其

KillExecutor

。worker在收到

KillExecutor

后调用

ExecutorRunner

的

kill()

函数：

case KillExecutor(appId, execId) =>

val fullId = appId + "/" + execId

executors.get(fullId) match {

case Some(executor) =>

logInfo("Asked to kill executor " + fullId)

executor.kill()

case None =>

logInfo("Asked to kill unknown executor " + fullId)

}

在

ExecutorRunner

内部，它会结束监控线程，同时结束监控线程所启动的进程，并且向worker汇报

ExecutorStateChanged

：

def kill() {

if (workerThread != null) {

workerThread.interrupt()

workerThread = null

if (process != null) {

logInfo("Killing process!")

process.destroy()

process.waitFor()

}

worker ! ExecutorStateChanged(appId, execId, ExecutorState.KILLED, None, None)

Runtime.getRuntime.removeShutdownHook(shutdownHook)

}

}

Application结束的同时清理了master和worker上的关于该application的所有信息，这样关于application结束的整个流程就介绍完了，当然在这里我们对于许多异常处理分支没有细究，但这并不影响我们对主线的把握。

End

至此对于deploy模块的分析暂告一个段落。deploy模块相对来说比较简单，也没有特别复杂的逻辑结构，正如前面所说的deploy模块是为了能让更多的没有部署Mesos的集群的用户能够使用Spark而实现的一种方案。
当然现阶段看来还略微简陋，比如application的调度方式(FIFO)是否会造成小应用长时间等待大应用的结束，是否有更好的调度策略；资源的衡量标准是否可以更多更合理，而不单单是cpu数量，因为现实场景中有的应用是disk intensive，有的是network intensive，这样就算cpu资源有富余，调度新的application也不一定会很有意义。
总的来说作为Mesos的一种简单替代方式，deploy模块对于推广Spark还是有积极意义的。