spark--度量系统

度量系统：在程序运行过程中对关键指标的一个记录。就是一个监控
Spark的度量系统使用codahale提供的第三方度量仓库Metrics。

1、代码结构：

source:指的是这些监控指标的来源

private[spark] trait Source {
  def sourceName: String
  def metricRegistry: MetricRegistry
}

private[master] class ApplicationSource(val application: ApplicationInfo) extends Source {
  override val metricRegistry = new MetricRegistry()
  override val sourceName = "%s.%s.%s".format("application", application.desc.name,
    System.currentTimeMillis())
  // 获取状态
  metricRegistry.register(MetricRegistry.name("status"), new Gauge[String] {
    override def getValue: String = application.state.toString
  })
 // 获取运行时间
  metricRegistry.register(MetricRegistry.name("runtime_ms"), new Gauge[Long] {
    override def getValue: Long = application.duration
  })
 // 获取cours数量
  metricRegistry.register(MetricRegistry.name("cores"), new Gauge[Int] {
    override def getValue: Int = application.coresGranted
  })

}

接口类定义名称，和MetricRegistry。

都类似于ApplicationSource，指标的来源

sink：指的是将source的指标输出的地方

private[spark] trait Sink {
  def start(): Unit
  def stop(): Unit
  def report(): Unit
}
private[spark] class ConsoleSink(val property: Properties, val registry: MetricRegistry,
    securityMgr: SecurityManager) extends Sink {
  val CONSOLE_DEFAULT_PERIOD = 10
  val CONSOLE_DEFAULT_UNIT = "SECONDS"

  val CONSOLE_KEY_PERIOD = "period"
  val CONSOLE_KEY_UNIT = "unit"

  val pollPeriod = Option(property.getProperty(CONSOLE_KEY_PERIOD)) match {
    case Some(s) => s.toInt
    case None => CONSOLE_DEFAULT_PERIOD
  }

  val pollUnit: TimeUnit = Option(property.getProperty(CONSOLE_KEY_UNIT)) match {
    case Some(s) => TimeUnit.valueOf(s.toUpperCase(Locale.ROOT))
    case None => TimeUnit.valueOf(CONSOLE_DEFAULT_UNIT)
  }

  MetricsSystem.checkMinimalPollingPeriod(pollUnit, pollPeriod)

  val reporter: ConsoleReporter = ConsoleReporter.forRegistry(registry)
      .convertDurationsTo(TimeUnit.MILLISECONDS)
      .convertRatesTo(TimeUnit.SECONDS)
      .build()

  override def start() {
    reporter.start(pollPeriod, pollUnit)
  }

  override def stop() {
    reporter.stop()
  }

  override def report() {
    reporter.report()
  }
}

MetricSystem: 对外的一个服务，可以用来注册source，sink，以及sink的启动

private[spark] class MetricsSystem private (
    val instance: String,
    conf: SparkConf,
    securityMgr: SecurityManager)
  extends Logging {

  private[this] val metricsConfig = new MetricsConfig(conf)

  private val sinks = new mutable.ArrayBuffer[Sink]
  private val sources = new mutable.ArrayBuffer[Source]
  private val registry = new MetricRegistry()

  private var running: Boolean = false

  // Treat MetricsServlet as a special sink as it should be exposed to add handlers to web ui
  private var metricsServlet: Option[MetricsServlet] = None

  /**
   * Get any UI handlers used by this metrics system; can only be called after start().
   */
  def getServletHandlers: Array[ServletContextHandler] = {
    require(running, "Can only call getServletHandlers on a running MetricsSystem")
    metricsServlet.map(_.getHandlers(conf)).getOrElse(Array())
  }

  metricsConfig.initialize()

  def start() {
    require(!running, "Attempting to start a MetricsSystem that is already running")
    running = true
    StaticSources.allSources.foreach(registerSource)
    registerSources()
    registerSinks()
    sinks.foreach(_.start)
  }

  def stop() {
    if (running) {
      sinks.foreach(_.stop)
    } else {
      logWarning("Stopping a MetricsSystem that is not running")
    }
    running = false
  }

  def report() {
    sinks.foreach(_.report())
  }

  /**
   * Build a name that uniquely identifies each metric source.
   * The name is structured as follows: <app ID>.<executor ID (or "driver")>.<source name>.
   * If either ID is not available, this defaults to just using <source name>.
   *
   * @param source Metric source to be named by this method.
   * @return An unique metric name for each combination of
   *         application, executor/driver and metric source.
   */
  private[spark] def buildRegistryName(source: Source): String = {
    val metricsNamespace = conf.get(METRICS_NAMESPACE).orElse(conf.getOption("spark.app.id"))

    val executorId = conf.getOption("spark.executor.id")
    val defaultName = MetricRegistry.name(source.sourceName)

    if (instance == "driver" || instance == "executor") {
      if (metricsNamespace.isDefined && executorId.isDefined) {
        MetricRegistry.name(metricsNamespace.get, executorId.get, source.sourceName)
      } else {
        // Only Driver and Executor set spark.app.id and spark.executor.id.
        // Other instance types, e.g. Master and Worker, are not related to a specific application.
        if (metricsNamespace.isEmpty) {
          logWarning(s"Using default name $defaultName for source because neither " +
            s"${METRICS_NAMESPACE.key} nor spark.app.id is set.")
        }
        if (executorId.isEmpty) {
          logWarning(s"Using default name $defaultName for source because spark.executor.id is " +
            s"not set.")
        }
        defaultName
      }
    } else { defaultName }
  }

  def getSourcesByName(sourceName: String): Seq[Source] =
    sources.filter(_.sourceName == sourceName)

  def registerSource(source: Source) {
    sources += source
    try {
      val regName = buildRegistryName(source)
      registry.register(regName, source.metricRegistry)
    } catch {
      case e: IllegalArgumentException => logInfo("Metrics already registered", e)
    }
  }

  def removeSource(source: Source) {
    sources -= source
    val regName = buildRegistryName(source)
    registry.removeMatching(new MetricFilter {
      def matches(name: String, metric: Metric): Boolean = name.startsWith(regName)
    })
  }
 
  private def registerSources() {
    val instConfig = metricsConfig.getInstance(instance)
    val sourceConfigs = metricsConfig.subProperties(instConfig, MetricsSystem.SOURCE_REGEX)

    // Register all the sources related to instance
    sourceConfigs.foreach { kv =>
      val classPath = kv._2.getProperty("class")
      try {
        val source = Utils.classForName(classPath).newInstance()
        registerSource(source.asInstanceOf[Source])
      } catch {
        case e: Exception => logError("Source class " + classPath + " cannot be instantiated", e)
      }
    }
  }

  private def registerSinks() {
    val instConfig = metricsConfig.getInstance(instance)
    val sinkConfigs = metricsConfig.subProperties(instConfig, MetricsSystem.SINK_REGEX)

    sinkConfigs.foreach { kv =>
      val classPath = kv._2.getProperty("class")
      if (null != classPath) {
        try {
          val sink = Utils.classForName(classPath)
            .getConstructor(classOf[Properties], classOf[MetricRegistry], classOf[SecurityManager])
            .newInstance(kv._2, registry, securityMgr)
          if (kv._1 == "servlet") {
            metricsServlet = Some(sink.asInstanceOf[MetricsServlet])
          } else {
            sinks += sink.asInstanceOf[Sink]
          }
        } catch {
          case e: Exception =>
            logError("Sink class " + classPath + " cannot be instantiated")
            throw e
        }
      }
    }
  }
}

private[spark] object MetricsSystem {
  val SINK_REGEX = "^sink\\.(.+)\\.(.+)".r
  val SOURCE_REGEX = "^source\\.(.+)\\.(.+)".r

  private[this] val MINIMAL_POLL_UNIT = TimeUnit.SECONDS
  private[this] val MINIMAL_POLL_PERIOD = 1

  def checkMinimalPollingPeriod(pollUnit: TimeUnit, pollPeriod: Int) {
    val period = MINIMAL_POLL_UNIT.convert(pollPeriod, pollUnit)
    if (period < MINIMAL_POLL_PERIOD) {
      throw new IllegalArgumentException("Polling period " + pollPeriod + " " + pollUnit +
        " below than minimal polling period ")
    }
  }

  def createMetricsSystem(
      instance: String, conf: SparkConf, securityMgr: SecurityManager): MetricsSystem = {
    new MetricsSystem(instance, conf, securityMgr)
  }
}

举个例子：比如一个application的状态监控，Master类会先创建一个metricsystem,然后注册一个ApplicationSource,然后启动这个监控

  private val applicationMetricsSystem = MetricsSystem.createMetricsSystem("applications", conf,
    securityMgr)
   applicationMetricsSystem.registerSource(app.appSource)
   applicationMetricsSystem.start()

相当于启动了一个consolereport去记录这个metrics