path: root/core/src/main/scala/org/apache/spark/SparkContext.scala

/*
 * Licensed to the Apache Software Foundation (ASF) under one or more
 * contributor license agreements.  See the NOTICE file distributed with
 * this work for additional information regarding copyright ownership.
 * The ASF licenses this file to You under the Apache License, Version 2.0
 * (the "License"); you may not use this file except in compliance with
 * the License.  You may obtain a copy of the License at
 *
 *    http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

package org.apache.spark

import java.io._
import java.net.URI
import java.util.{Properties, UUID}
import java.util.concurrent.atomic.AtomicInteger

import scala.collection.{Map, Set}
import scala.collection.generic.Growable
import scala.collection.mutable.{ArrayBuffer, HashMap}
import scala.reflect.{ClassTag, classTag}

import org.apache.hadoop.conf.Configuration
import org.apache.hadoop.fs.Path
import org.apache.hadoop.io.{ArrayWritable, BooleanWritable, BytesWritable, DoubleWritable, FloatWritable, IntWritable, LongWritable, NullWritable, Text, Writable}
import org.apache.hadoop.mapred.{FileInputFormat, InputFormat, JobConf, SequenceFileInputFormat, TextInputFormat}
import org.apache.hadoop.mapreduce.{InputFormat => NewInputFormat, Job => NewHadoopJob}
import org.apache.hadoop.mapreduce.lib.input.{FileInputFormat => NewFileInputFormat}
import org.apache.mesos.MesosNativeLibrary

import org.apache.spark.deploy.{LocalSparkCluster, SparkHadoopUtil}
import org.apache.spark.partial.{ApproximateEvaluator, PartialResult}
import org.apache.spark.rdd._
import org.apache.spark.scheduler._
import org.apache.spark.scheduler.cluster.{CoarseGrainedSchedulerBackend, SparkDeploySchedulerBackend, SimrSchedulerBackend}
import org.apache.spark.scheduler.cluster.mesos.{CoarseMesosSchedulerBackend, MesosSchedulerBackend}
import org.apache.spark.scheduler.local.LocalBackend
import org.apache.spark.storage.{BlockManagerSource, RDDInfo, StorageStatus, StorageUtils}
import org.apache.spark.ui.SparkUI
import org.apache.spark.util.{ClosureCleaner, MetadataCleaner, MetadataCleanerType, TimeStampedHashMap, Utils}

/**
 * Main entry point for Spark functionality. A SparkContext represents the connection to a Spark
 * cluster, and can be used to create RDDs, accumulators and broadcast variables on that cluster.
 *
 * @param config a Spark Config object describing the application configuration. Any settings in
 *   this config overrides the default configs as well as system properties.
 * @param preferredNodeLocationData used in YARN mode to select nodes to launch containers on. Can
 *   be generated using [[org.apache.spark.scheduler.InputFormatInfo.computePreferredLocations]]
 *   from a list of input files or InputFormats for the application.
 */
class SparkContext(
    config: SparkConf,
    // This is used only by YARN for now, but should be relevant to other cluster types (Mesos,
    // etc) too. This is typically generated from InputFormatInfo.computePreferredLocations. It
    // contains a map from hostname to a list of input format splits on the host.
    val preferredNodeLocationData: Map[String, Set[SplitInfo]] = Map())
  extends Logging {

  /**
   * Alternative constructor that allows setting common Spark properties directly
   *
   * @param master Cluster URL to connect to (e.g. mesos://host:port, spark://host:port, local[4]).
   * @param appName A name for your application, to display on the cluster web UI
   * @param conf a [[org.apache.spark.SparkConf]] object specifying other Spark parameters
   */
  def this(master: String, appName: String, conf: SparkConf) =
    this(SparkContext.updatedConf(conf, master, appName))

  /**
   * Alternative constructor that allows setting common Spark properties directly
   *
   * @param master Cluster URL to connect to (e.g. mesos://host:port, spark://host:port, local[4]).
   * @param appName A name for your application, to display on the cluster web UI.
   * @param sparkHome Location where Spark is installed on cluster nodes.
   * @param jars Collection of JARs to send to the cluster. These can be paths on the local file
   *             system or HDFS, HTTP, HTTPS, or FTP URLs.
   * @param environment Environment variables to set on worker nodes.
   */
  def this(
      master: String,
      appName: String,
      sparkHome: String = null,
      jars: Seq[String] = Nil,
      environment: Map[String, String] = Map(),
      preferredNodeLocationData: Map[String, Set[SplitInfo]] = Map()) =
  {
    this(SparkContext.updatedConf(new SparkConf(), master, appName, sparkHome, jars, environment),
      preferredNodeLocationData)
  }

  private[spark] val conf = config.clone()

  /**
   * Return a copy of this SparkContext's configuration. The configuration ''cannot'' be
   * changed at runtime.
   */
  def getConf: SparkConf = conf.clone()

  if (!conf.contains("spark.master")) {
    throw new SparkException("A master URL must be set in your configuration")
  }
  if (!conf.contains("spark.app.name")) {
    throw new SparkException("An application must be set in your configuration")
  }

  if (conf.getBoolean("spark.logConf", false)) {
    logInfo("Spark configuration:\n" + conf.toDebugString)
  }

  // Set Spark driver host and port system properties
  conf.setIfMissing("spark.driver.host", Utils.localHostName())
  conf.setIfMissing("spark.driver.port", "0")

  val jars: Seq[String] = if (conf.contains("spark.jars")) {
    conf.get("spark.jars").split(",").filter(_.size != 0)
  } else {
    null
  }

  val master = conf.get("spark.master")
  val appName = conf.get("spark.app.name")

  val isLocal = (master == "local" || master.startsWith("local["))

  if (master == "yarn-client") System.setProperty("SPARK_YARN_MODE", "true")

  // Create the Spark execution environment (cache, map output tracker, etc)
  private[spark] val env = SparkEnv.create(
    conf,
    "<driver>",
    conf.get("spark.driver.host"),
    conf.get("spark.driver.port").toInt,
    isDriver = true,
    isLocal = isLocal)
  SparkEnv.set(env)

  // Used to store a URL for each static file/jar together with the file's local timestamp
  private[spark] val addedFiles = HashMap[String, Long]()
  private[spark] val addedJars = HashMap[String, Long]()

  // Keeps track of all persisted RDDs
  private[spark] val persistentRdds = new TimeStampedHashMap[Int, RDD[_]]
  private[spark] val metadataCleaner =
    new MetadataCleaner(MetadataCleanerType.SPARK_CONTEXT, this.cleanup, conf)

  // Initialize the Spark UI
  private[spark] val ui = new SparkUI(this)
  ui.bind()

  val startTime = System.currentTimeMillis()

  // Add each JAR given through the constructor
  if (jars != null) {
    jars.foreach(addJar)
  }

  private[spark] val executorMemory = conf.getOption("spark.executor.memory")
    .orElse(Option(System.getenv("SPARK_MEM")))
    .map(Utils.memoryStringToMb)
    .getOrElse(512)

  if (!conf.contains("spark.executor.memory") && sys.env.contains("SPARK_MEM")) {
    logWarning("Using SPARK_MEM to set amount of memory to use per executor process is " +
      "deprecated, instead use spark.executor.memory")
  }

  // Environment variables to pass to our executors
  private[spark] val executorEnvs = HashMap[String, String]()
  // Note: SPARK_MEM is included for Mesos, but overwritten for standalone mode in ExecutorRunner
  for (key <- Seq("SPARK_CLASSPATH", "SPARK_LIBRARY_PATH", "SPARK_JAVA_OPTS");
      value <- Option(System.getenv(key))) {
    executorEnvs(key) = value
  }
  // Convert java options to env vars as a work around
  // since we can't set env vars directly in sbt.
  for { (envKey, propKey) <- Seq(("SPARK_HOME", "spark.home"), ("SPARK_TESTING", "spark.testing"))
    value <- Option(System.getenv(envKey)).orElse(Option(System.getProperty(propKey)))} {
    executorEnvs(envKey) = value
  }
  // Since memory can be set with a system property too, use that
  executorEnvs("SPARK_MEM") = executorMemory + "m"
  executorEnvs ++= conf.getExecutorEnv

  // Set SPARK_USER for user who is running SparkContext.
  val sparkUser = Option {
    Option(System.getProperty("user.name")).getOrElse(System.getenv("SPARK_USER"))
  }.getOrElse {
    SparkContext.SPARK_UNKNOWN_USER
  }
  executorEnvs("SPARK_USER") = sparkUser

  // Create and start the scheduler
  private[spark] var taskScheduler = SparkContext.createTaskScheduler(this, master, appName)
  taskScheduler.start()

  @volatile private[spark] var dagScheduler = new DAGScheduler(taskScheduler)
  dagScheduler.start()

  ui.start()

  /** A default Hadoop Configuration for the Hadoop code (e.g. file systems) that we reuse. */
  val hadoopConfiguration = {
    val env = SparkEnv.get
    val hadoopConf = SparkHadoopUtil.get.newConfiguration()
    // Explicitly check for S3 environment variables
    if (System.getenv("AWS_ACCESS_KEY_ID") != null &&
        System.getenv("AWS_SECRET_ACCESS_KEY") != null) {
      hadoopConf.set("fs.s3.awsAccessKeyId", System.getenv("AWS_ACCESS_KEY_ID"))
      hadoopConf.set("fs.s3n.awsAccessKeyId", System.getenv("AWS_ACCESS_KEY_ID"))
      hadoopConf.set("fs.s3.awsSecretAccessKey", System.getenv("AWS_SECRET_ACCESS_KEY"))
      hadoopConf.set("fs.s3n.awsSecretAccessKey", System.getenv("AWS_SECRET_ACCESS_KEY"))
    }
    // Copy any "spark.hadoop.foo=bar" system properties into conf as "foo=bar"
    conf.getAll.foreach { case (key, value) =>
      if (key.startsWith("spark.hadoop.")) {
        hadoopConf.set(key.substring("spark.hadoop.".length), value)
      }
    }
    val bufferSize = conf.get("spark.buffer.size", "65536")
    hadoopConf.set("io.file.buffer.size", bufferSize)
    hadoopConf
  }

  private[spark] var checkpointDir: Option[String] = None

  // Thread Local variable that can be used by users to pass information down the stack
  private val localProperties = new InheritableThreadLocal[Properties] {
    override protected def childValue(parent: Properties): Properties = new Properties(parent)
  }

  private[spark] def getLocalProperties: Properties = localProperties.get()

  private[spark] def setLocalProperties(props: Properties) {
    localProperties.set(props)
  }

  @deprecated("Properties no longer need to be explicitly initialized.", "1.0.0")
  def initLocalProperties() {
    localProperties.set(new Properties())
  }

  /**
   * Set a local property that affects jobs submitted from this thread, such as the
   * Spark fair scheduler pool.
   */
  def setLocalProperty(key: String, value: String) {
    if (localProperties.get() == null) {
      localProperties.set(new Properties())
    }
    if (value == null) {
      localProperties.get.remove(key)
    } else {
      localProperties.get.setProperty(key, value)
    }
  }

  /**
   * Get a local property set in this thread, or null if it is missing. See
   * [[org.apache.spark.SparkContext.setLocalProperty]].
   */
  def getLocalProperty(key: String): String =
    Option(localProperties.get).map(_.getProperty(key)).getOrElse(null)

  /** Set a human readable description of the current job. */
  @deprecated("use setJobGroup", "0.8.1")
  def setJobDescription(value: String) {
    setLocalProperty(SparkContext.SPARK_JOB_DESCRIPTION, value)
  }

  /**
   * Assigns a group ID to all the jobs started by this thread until the group ID is set to a
   * different value or cleared.
   *
   * Often, a unit of execution in an application consists of multiple Spark actions or jobs.
   * Application programmers can use this method to group all those jobs together and give a
   * group description. Once set, the Spark web UI will associate such jobs with this group.
   *
   * The application can also use [[org.apache.spark.SparkContext.cancelJobGroup]] to cancel all
   * running jobs in this group. For example,
   * {{{
   * // In the main thread:
   * sc.setJobGroup("some_job_to_cancel", "some job description")
   * sc.parallelize(1 to 10000, 2).map { i => Thread.sleep(10); i }.count()
   *
   * // In a separate thread:
   * sc.cancelJobGroup("some_job_to_cancel")
   * }}}
   */
  def setJobGroup(groupId: String, description: String) {
    setLocalProperty(SparkContext.SPARK_JOB_DESCRIPTION, description)
    setLocalProperty(SparkContext.SPARK_JOB_GROUP_ID, groupId)
  }

  /** Clear the current thread's job group ID and its description. */
  def clearJobGroup() {
    setLocalProperty(SparkContext.SPARK_JOB_DESCRIPTION, null)
    setLocalProperty(SparkContext.SPARK_JOB_GROUP_ID, null)
  }

  // Post init
  taskScheduler.postStartHook()

  private val dagSchedulerSource = new DAGSchedulerSource(this.dagScheduler, this)
  private val blockManagerSource = new BlockManagerSource(SparkEnv.get.blockManager, this)

  private def initDriverMetrics() {
    SparkEnv.get.metricsSystem.registerSource(dagSchedulerSource)
    SparkEnv.get.metricsSystem.registerSource(blockManagerSource)
  }

  initDriverMetrics()

  // Methods for creating RDDs

  /** Distribute a local Scala collection to form an RDD. */
  def parallelize[T: ClassTag](seq: Seq[T], numSlices: Int = defaultParallelism): RDD[T] = {
    new ParallelCollectionRDD[T](this, seq, numSlices, Map[Int, Seq[String]]())
  }

  /** Distribute a local Scala collection to form an RDD. */
  def makeRDD[T: ClassTag](seq: Seq[T], numSlices: Int = defaultParallelism): RDD[T] = {
    parallelize(seq, numSlices)
  }

  /** Distribute a local Scala collection to form an RDD, with one or more
    * location preferences (hostnames of Spark nodes) for each object.
    * Create a new partition for each collection item. */
   def makeRDD[T: ClassTag](seq: Seq[(T, Seq[String])]): RDD[T] = {
    val indexToPrefs = seq.zipWithIndex.map(t => (t._2, t._1._2)).toMap
    new ParallelCollectionRDD[T](this, seq.map(_._1), seq.size, indexToPrefs)
  }

  /**
   * Read a text file from HDFS, a local file system (available on all nodes), or any
   * Hadoop-supported file system URI, and return it as an RDD of Strings.
   */
  def textFile(path: String, minSplits: Int = defaultMinSplits): RDD[String] = {
    hadoopFile(path, classOf[TextInputFormat], classOf[LongWritable], classOf[Text],
      minSplits).map(pair => pair._2.toString)
  }

  /**
   * Get an RDD for a Hadoop-readable dataset from a Hadoop JobConf given its InputFormat and other
   * necessary info (e.g. file name for a filesystem-based dataset, table name for HyperTable),
   * using the older MapReduce API (`org.apache.hadoop.mapred`).
   *
   * @param conf JobConf for setting up the dataset
   * @param inputFormatClass Class of the InputFormat
   * @param keyClass Class of the keys
   * @param valueClass Class of the values
   * @param minSplits Minimum number of Hadoop Splits to generate.
   *
   * '''Note:''' Because Hadoop's RecordReader class re-uses the same Writable object for each
   * record, directly caching the returned RDD will create many references to the same object.
   * If you plan to directly cache Hadoop writable objects, you should first copy them using
   * a `map` function.
   */
  def hadoopRDD[K, V](
      conf: JobConf,
      inputFormatClass: Class[_ <: InputFormat[K, V]],
      keyClass: Class[K],
      valueClass: Class[V],
      minSplits: Int = defaultMinSplits
      ): RDD[(K, V)] = {
    // Add necessary security credentials to the JobConf before broadcasting it.
    SparkHadoopUtil.get.addCredentials(conf)
    new HadoopRDD(this, conf, inputFormatClass, keyClass, valueClass, minSplits)
  }

  /** Get an RDD for a Hadoop file with an arbitrary InputFormat
    *
    * '''Note:''' Because Hadoop's RecordReader class re-uses the same Writable object for each
    * record, directly caching the returned RDD will create many references to the same object.
    * If you plan to directly cache Hadoop writable objects, you should first copy them using
    * a `map` function.
    * */
  def hadoopFile[K, V](
      path: String,
      inputFormatClass: Class[_ <: InputFormat[K, V]],
      keyClass: Class[K],
      valueClass: Class[V],
      minSplits: Int = defaultMinSplits
      ): RDD[(K, V)] = {
    // A Hadoop configuration can be about 10 KB, which is pretty big, so broadcast it.
    val confBroadcast = broadcast(new SerializableWritable(hadoopConfiguration))
    val setInputPathsFunc = (jobConf: JobConf) => FileInputFormat.setInputPaths(jobConf, path)
    new HadoopRDD(
      this,
      confBroadcast,
      Some(setInputPathsFunc),
      inputFormatClass,
      keyClass,
      valueClass,
      minSplits)
  }

  /**
   * Smarter version of hadoopFile() that uses class tags to figure out the classes of keys,
   * values and the InputFormat so that users don't need to pass them directly. Instead, callers
   * can just write, for example,
   * {{{
   * val file = sparkContext.hadoopFile[LongWritable, Text, TextInputFormat](path, minSplits)
   * }}}
   *
   * '''Note:''' Because Hadoop's RecordReader class re-uses the same Writable object for each
   * record, directly caching the returned RDD will create many references to the same object.
   * If you plan to directly cache Hadoop writable objects, you should first copy them using
   * a `map` function.
   */
  def hadoopFile[K, V, F <: InputFormat[K, V]]
      (path: String, minSplits: Int)
      (implicit km: ClassTag[K], vm: ClassTag[V], fm: ClassTag[F]): RDD[(K, V)] = {
    hadoopFile(path,
      fm.runtimeClass.asInstanceOf[Class[F]],
      km.runtimeClass.asInstanceOf[Class[K]],
      vm.runtimeClass.asInstanceOf[Class[V]],
      minSplits)
  }

  /**
   * Smarter version of hadoopFile() that uses class tags to figure out the classes of keys,
   * values and the InputFormat so that users don't need to pass them directly. Instead, callers
   * can just write, for example,
   * {{{
   * val file = sparkContext.hadoopFile[LongWritable, Text, TextInputFormat](path)
   * }}}
   *
   * '''Note:''' Because Hadoop's RecordReader class re-uses the same Writable object for each
   * record, directly caching the returned RDD will create many references to the same object.
   * If you plan to directly cache Hadoop writable objects, you should first copy them using
   * a `map` function.
   */
  def hadoopFile[K, V, F <: InputFormat[K, V]](path: String)
      (implicit km: ClassTag[K], vm: ClassTag[V], fm: ClassTag[F]): RDD[(K, V)] =
    hadoopFile[K, V, F](path, defaultMinSplits)

  /** Get an RDD for a Hadoop file with an arbitrary new API InputFormat. */
  def newAPIHadoopFile[K, V, F <: NewInputFormat[K, V]]
      (path: String)
      (implicit km: ClassTag[K], vm: ClassTag[V], fm: ClassTag[F]): RDD[(K, V)] = {
    newAPIHadoopFile(
      path,
      fm.runtimeClass.asInstanceOf[Class[F]],
      km.runtimeClass.asInstanceOf[Class[K]],
      vm.runtimeClass.asInstanceOf[Class[V]])
  }

  /**
   * Get an RDD for a given Hadoop file with an arbitrary new API InputFormat
   * and extra configuration options to pass to the input format.
   *
   * '''Note:''' Because Hadoop's RecordReader class re-uses the same Writable object for each
   * record, directly caching the returned RDD will create many references to the same object.
   * If you plan to directly cache Hadoop writable objects, you should first copy them using
   * a `map` function.
   */
  def newAPIHadoopFile[K, V, F <: NewInputFormat[K, V]](
      path: String,
      fClass: Class[F],
      kClass: Class[K],
      vClass: Class[V],
      conf: Configuration = hadoopConfiguration): RDD[(K, V)] = {
    val job = new NewHadoopJob(conf)
    NewFileInputFormat.addInputPath(job, new Path(path))
    val updatedConf = job.getConfiguration
    new NewHadoopRDD(this, fClass, kClass, vClass, updatedConf)
  }

  /**
   * Get an RDD for a given Hadoop file with an arbitrary new API InputFormat
   * and extra configuration options to pass to the input format.
   *
   * '''Note:''' Because Hadoop's RecordReader class re-uses the same Writable object for each
   * record, directly caching the returned RDD will create many references to the same object.
   * If you plan to directly cache Hadoop writable objects, you should first copy them using
   * a `map` function.
   */
  def newAPIHadoopRDD[K, V, F <: NewInputFormat[K, V]](
      conf: Configuration = hadoopConfiguration,
      fClass: Class[F],
      kClass: Class[K],
      vClass: Class[V]): RDD[(K, V)] = {
    new NewHadoopRDD(this, fClass, kClass, vClass, conf)
  }

  /** Get an RDD for a Hadoop SequenceFile with given key and value types.
    *
    * '''Note:''' Because Hadoop's RecordReader class re-uses the same Writable object for each
    * record, directly caching the returned RDD will create many references to the same object.
    * If you plan to directly cache Hadoop writable objects, you should first copy them using
    * a `map` function.
    */
  def sequenceFile[K, V](path: String,
      keyClass: Class[K],
      valueClass: Class[V],
      minSplits: Int
      ): RDD[(K, V)] = {
    val inputFormatClass = classOf[SequenceFileInputFormat[K, V]]
    hadoopFile(path, inputFormatClass, keyClass, valueClass, minSplits)
  }

  /** Get an RDD for a Hadoop SequenceFile with given key and value types.
    *
    * '''Note:''' Because Hadoop's RecordReader class re-uses the same Writable object for each
    * record, directly caching the returned RDD will create many references to the same object.
    * If you plan to directly cache Hadoop writable objects, you should first copy them using
    * a `map` function.
    * */
  def sequenceFile[K, V](path: String, keyClass: Class[K], valueClass: Class[V]
      ): RDD[(K, V)] =
    sequenceFile(path, keyClass, valueClass, defaultMinSplits)

  /**
   * Version of sequenceFile() for types implicitly convertible to Writables through a
   * WritableConverter. For example, to access a SequenceFile where the keys are Text and the
   * values are IntWritable, you could simply write
   * {{{
   * sparkContext.sequenceFile[String, Int](path, ...)
   * }}}
   *
   * WritableConverters are provided in a somewhat strange way (by an implicit function) to support
   * both subclasses of Writable and types for which we define a converter (e.g. Int to
   * IntWritable). The most natural thing would've been to have implicit objects for the
   * converters, but then we couldn't have an object for every subclass of Writable (you can't
   * have a parameterized singleton object). We use functions instead to create a new converter
   * for the appropriate type. In addition, we pass the converter a ClassTag of its type to
   * allow it to figure out the Writable class to use in the subclass case.
   *
   * '''Note:''' Because Hadoop's RecordReader class re-uses the same Writable object for each
   * record, directly caching the returned RDD will create many references to the same object.
   * If you plan to directly cache Hadoop writable objects, you should first copy them using
   * a `map` function.
   */
   def sequenceFile[K, V]
       (path: String, minSplits: Int = defaultMinSplits)
       (implicit km: ClassTag[K], vm: ClassTag[V],
        kcf: () => WritableConverter[K], vcf: () => WritableConverter[V])
      : RDD[(K, V)] = {
    val kc = kcf()
    val vc = vcf()
    val format = classOf[SequenceFileInputFormat[Writable, Writable]]
    val writables = hadoopFile(path, format,
        kc.writableClass(km).asInstanceOf[Class[Writable]],
        vc.writableClass(vm).asInstanceOf[Class[Writable]], minSplits)
    writables.map { case (k, v) => (kc.convert(k), vc.convert(v)) }
  }

  /**
   * Load an RDD saved as a SequenceFile containing serialized objects, with NullWritable keys and
   * BytesWritable values that contain a serialized partition. This is still an experimental
   * storage format and may not be supported exactly as is in future Spark releases. It will also
   * be pretty slow if you use the default serializer (Java serialization),
   * though the nice thing about it is that there's very little effort required to save arbitrary
   * objects.
   */
  def objectFile[T: ClassTag](
      path: String,
      minSplits: Int = defaultMinSplits
      ): RDD[T] = {
    sequenceFile(path, classOf[NullWritable], classOf[BytesWritable], minSplits)
      .flatMap(x => Utils.deserialize[Array[T]](x._2.getBytes))
  }


  protected[spark] def checkpointFile[T: ClassTag](
      path: String
    ): RDD[T] = {
    new CheckpointRDD[T](this, path)
  }

  /** Build the union of a list of RDDs. */
  def union[T: ClassTag](rdds: Seq[RDD[T]]): RDD[T] = new UnionRDD(this, rdds)

  /** Build the union of a list of RDDs passed as variable-length arguments. */
  def union[T: ClassTag](first: RDD[T], rest: RDD[T]*): RDD[T] =
    new UnionRDD(this, Seq(first) ++ rest)

  // Methods for creating shared variables

  /**
   * Create an [[org.apache.spark.Accumulator]] variable of a given type, which tasks can "add"
   * values to using the `+=` method. Only the driver can access the accumulator's `value`.
   */
  def accumulator[T](initialValue: T)(implicit param: AccumulatorParam[T]) =
    new Accumulator(initialValue, param)

  /**
   * Create an [[org.apache.spark.Accumulable]] shared variable, to which tasks can add values
   * with `+=`. Only the driver can access the accumuable's `value`.
   * @tparam T accumulator type
   * @tparam R type that can be added to the accumulator
   */
  def accumulable[T, R](initialValue: T)(implicit param: AccumulableParam[T, R]) =
    new Accumulable(initialValue, param)

  /**
   * Create an accumulator from a "mutable collection" type.
   *
   * Growable and TraversableOnce are the standard APIs that guarantee += and ++=, implemented by
   * standard mutable collections. So you can use this with mutable Map, Set, etc.
   */
  def accumulableCollection[R <% Growable[T] with TraversableOnce[T] with Serializable, T]
      (initialValue: R) = {
    val param = new GrowableAccumulableParam[R,T]
    new Accumulable(initialValue, param)
  }

  /**
   * Broadcast a read-only variable to the cluster, returning a
   * [[org.apache.spark.broadcast.Broadcast]] object for reading it in distributed functions.
   * The variable will be sent to each cluster only once.
   */
  def broadcast[T](value: T) = env.broadcastManager.newBroadcast[T](value, isLocal)

  /**
   * Add a file to be downloaded with this Spark job on every node.
   * The `path` passed can be either a local file, a file in HDFS (or other Hadoop-supported
   * filesystems), or an HTTP, HTTPS or FTP URI.  To access the file in Spark jobs,
   * use `SparkFiles.get(path)` to find its download location.
   */
  def addFile(path: String) {
    val uri = new URI(path)
    val key = uri.getScheme match {
      case null | "file" => env.httpFileServer.addFile(new File(uri.getPath))
      case "local"       => "file:" + uri.getPath
      case _             => path
    }
    addedFiles(key) = System.currentTimeMillis

    // Fetch the file locally in case a job is executed using DAGScheduler.runLocally().
    Utils.fetchFile(path, new File(SparkFiles.getRootDirectory), conf, env.securityManager)

    logInfo("Added file " + path + " at " + key + " with timestamp " + addedFiles(key))
  }

  def addSparkListener(listener: SparkListener) {
    dagScheduler.addSparkListener(listener)
  }

  /**
   * Return a map from the slave to the max memory available for caching and the remaining
   * memory available for caching.
   */
  def getExecutorMemoryStatus: Map[String, (Long, Long)] = {
    env.blockManager.master.getMemoryStatus.map { case(blockManagerId, mem) =>
      (blockManagerId.host + ":" + blockManagerId.port, mem)
    }
  }

  /**
   * Return information about what RDDs are cached, if they are in mem or on disk, how much space
   * they take, etc.
   */
  def getRDDStorageInfo: Array[RDDInfo] = {
    StorageUtils.rddInfoFromStorageStatus(getExecutorStorageStatus, this)
  }

  /**
   * Returns an immutable map of RDDs that have marked themselves as persistent via cache() call.
   * Note that this does not necessarily mean the caching or computation was successful.
   */
  def getPersistentRDDs: Map[Int, RDD[_]] = persistentRdds.toMap

  def getStageInfo: Map[Stage,StageInfo] = {
    dagScheduler.stageToInfos
  }

  /**
   * Return information about blocks stored in all of the slaves
   */
  def getExecutorStorageStatus: Array[StorageStatus] = {
    env.blockManager.master.getStorageStatus
  }

  /**
   *  Return pools for fair scheduler
   *  TODO(xiajunluan): We should take nested pools into account
   */
  def getAllPools: ArrayBuffer[Schedulable] = {
    taskScheduler.rootPool.schedulableQueue
  }

  /**
   * Return the pool associated with the given name, if one exists
   */
  def getPoolForName(pool: String): Option[Schedulable] = {
    taskScheduler.rootPool.schedulableNameToSchedulable.get(pool)
  }

  /**
   *  Return current scheduling mode
   */
  def getSchedulingMode: SchedulingMode.SchedulingMode = {
    taskScheduler.schedulingMode
  }

  /**
   * Clear the job's list of files added by `addFile` so that they do not get downloaded to
   * any new nodes.
   */
  def clearFiles() {
    addedFiles.clear()
  }

  /**
   * Gets the locality information associated with the partition in a particular rdd
   * @param rdd of interest
   * @param partition to be looked up for locality
   * @return list of preferred locations for the partition
   */
  private [spark] def getPreferredLocs(rdd: RDD[_], partition: Int): Seq[TaskLocation] = {
    dagScheduler.getPreferredLocs(rdd, partition)
  }

  /**
   * Adds a JAR dependency for all tasks to be executed on this SparkContext in the future.
   * The `path` passed can be either a local file, a file in HDFS (or other Hadoop-supported
   * filesystems), an HTTP, HTTPS or FTP URI, or local:/path for a file on every worker node.
   */
  def addJar(path: String) {
    if (path == null) {
      logWarning("null specified as parameter to addJar")
    } else {
      var key = ""
      if (path.contains("\\")) {
        // For local paths with backslashes on Windows, URI throws an exception
        key = env.httpFileServer.addJar(new File(path))
      } else {
        val uri = new URI(path)
        key = uri.getScheme match {
          // A JAR file which exists only on the driver node
          case null | "file" =>
            if (SparkHadoopUtil.get.isYarnMode() && master == "yarn-standalone") {
              // In order for this to work in yarn standalone mode the user must specify the 
              // --addjars option to the client to upload the file into the distributed cache 
              // of the AM to make it show up in the current working directory.
              val fileName = new Path(uri.getPath).getName()
              try {
                env.httpFileServer.addJar(new File(fileName))
              } catch {
                case e: Exception => {
                  // For now just log an error but allow to go through so spark examples work.
                  // The spark examples don't really need the jar distributed since its also 
                  // the app jar.
                  logError("Error adding jar (" + e + "), was the --addJars option used?")
                  null
                }
              }
            } else {
              env.httpFileServer.addJar(new File(uri.getPath))
            }
          // A JAR file which exists locally on every worker node
          case "local" =>
            "file:" + uri.getPath
          case _ =>
            path
        }
      }
      if (key != null) {
        addedJars(key) = System.currentTimeMillis
        logInfo("Added JAR " + path + " at " + key + " with timestamp " + addedJars(key))
      }
    }
  }

  /**
   * Clear the job's list of JARs added by `addJar` so that they do not get downloaded to
   * any new nodes.
   */
  def clearJars() {
    addedJars.clear()
  }

  /** Shut down the SparkContext. */
  def stop() {
    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()
      dagSchedulerCopy.stop()
      taskScheduler = null
      // TODO: Cache.stop()?
      env.stop()
      // Clean up locally linked files
      clearFiles()
      clearJars()
      SparkEnv.set(null)
      ShuffleMapTask.clearCache()
      ResultTask.clearCache()
      logInfo("Successfully stopped SparkContext")
    } else {
      logInfo("SparkContext already stopped")
    }
  }


  /**
   * Get Spark's home location from either a value set through the constructor,
   * or the spark.home Java property, or the SPARK_HOME environment variable
   * (in that order of preference). If neither of these is set, return None.
   */
  private[spark] def getSparkHome(): Option[String] = {
    conf.getOption("spark.home").orElse(Option(System.getenv("SPARK_HOME")))
  }

  /**
   * Support function for API backtraces.
   */
  def setCallSite(site: String) {
    setLocalProperty("externalCallSite", site)
  }

  /**
   * Support function for API backtraces.
   */
  def clearCallSite() {
    setLocalProperty("externalCallSite", null)
  }

  private[spark] def getCallSite(): String = {
    val callSite = getLocalProperty("externalCallSite")
    if (callSite == null) {
      Utils.formatSparkCallSite
    } else {
      callSite
    }
  }

  /**
   * Run a function on a given set of partitions in an RDD and pass the results to the given
   * handler function. This is the main entry point for all actions in Spark. The allowLocal
   * flag specifies whether the scheduler can run the computation on the driver rather than
   * shipping it out to the cluster, for short actions like first().
   */
  def runJob[T, U: ClassTag](
      rdd: RDD[T],
      func: (TaskContext, Iterator[T]) => U,
      partitions: Seq[Int],
      allowLocal: Boolean,
      resultHandler: (Int, U) => Unit) {
    val callSite = getCallSite
    val cleanedFunc = clean(func)
    logInfo("Starting job: " + callSite)
    val start = System.nanoTime
    dagScheduler.runJob(rdd, cleanedFunc, partitions, callSite, allowLocal,
      resultHandler, localProperties.get)
    logInfo("Job finished: " + callSite + ", took " + (System.nanoTime - start) / 1e9 + " s")
    rdd.doCheckpoint()
  }

  /**
   * Run a function on a given set of partitions in an RDD and return the results as an array. The
   * allowLocal flag specifies whether the scheduler can run the computation on the driver rather
   * than shipping it out to the cluster, for short actions like first().
   */
  def runJob[T, U: ClassTag](
      rdd: RDD[T],
      func: (TaskContext, Iterator[T]) => U,
      partitions: Seq[Int],
      allowLocal: Boolean
      ): Array[U] = {
    val results = new Array[U](partitions.size)
    runJob[T, U](rdd, func, partitions, allowLocal, (index, res) => results(index) = res)
    results
  }

  /**
   * Run a job on a given set of partitions of an RDD, but take a function of type
   * `Iterator[T] => U` instead of `(TaskContext, Iterator[T]) => U`.
   */
  def runJob[T, U: ClassTag](
      rdd: RDD[T],
      func: Iterator[T] => U,
      partitions: Seq[Int],
      allowLocal: Boolean
      ): Array[U] = {
    runJob(rdd, (context: TaskContext, iter: Iterator[T]) => func(iter), partitions, allowLocal)
  }

  /**
   * Run a job on all partitions in an RDD and return the results in an array.
   */
  def runJob[T, U: ClassTag](rdd: RDD[T], func: (TaskContext, Iterator[T]) => U): Array[U] = {
    runJob(rdd, func, 0 until rdd.partitions.size, false)
  }

  /**
   * Run a job on all partitions in an RDD and return the results in an array.
   */
  def runJob[T, U: ClassTag](rdd: RDD[T], func: Iterator[T] => U): Array[U] = {
    runJob(rdd, func, 0 until rdd.partitions.size, false)
  }

  /**
   * Run a job on all partitions in an RDD and pass the results to a handler function.
   */
  def runJob[T, U: ClassTag](
    rdd: RDD[T],
    processPartition: (TaskContext, Iterator[T]) => U,
    resultHandler: (Int, U) => Unit)
  {
    runJob[T, U](rdd, processPartition, 0 until rdd.partitions.size, false, resultHandler)
  }

  /**
   * Run a job on all partitions in an RDD and pass the results to a handler function.
   */
  def runJob[T, U: ClassTag](
      rdd: RDD[T],
      processPartition: Iterator[T] => U,
      resultHandler: (Int, U) => Unit)
  {
    val processFunc = (context: TaskContext, iter: Iterator[T]) => processPartition(iter)
    runJob[T, U](rdd, processFunc, 0 until rdd.partitions.size, false, resultHandler)
  }

  /**
   * Run a job that can return approximate results.
   */
  def runApproximateJob[T, U, R](
      rdd: RDD[T],
      func: (TaskContext, Iterator[T]) => U,
      evaluator: ApproximateEvaluator[U, R],
      timeout: Long): PartialResult[R] = {
    val callSite = getCallSite
    logInfo("Starting job: " + callSite)
    val start = System.nanoTime
    val result = dagScheduler.runApproximateJob(rdd, func, evaluator, callSite, timeout,
      localProperties.get)
    logInfo("Job finished: " + callSite + ", took " + (System.nanoTime - start) / 1e9 + " s")
    result
  }

  /**
   * Submit a job for execution and return a FutureJob holding the result.
   */
  def submitJob[T, U, R](
      rdd: RDD[T],
      processPartition: Iterator[T] => U,
      partitions: Seq[Int],
      resultHandler: (Int, U) => Unit,
      resultFunc: => R): SimpleFutureAction[R] =
  {
    val cleanF = clean(processPartition)
    val callSite = getCallSite
    val waiter = dagScheduler.submitJob(
      rdd,
      (context: TaskContext, iter: Iterator[T]) => cleanF(iter),
      partitions,
      callSite,
      allowLocal = false,
      resultHandler,
      localProperties.get)
    new SimpleFutureAction(waiter, resultFunc)
  }

  /**
   * Cancel active jobs for the specified group. See [[org.apache.spark.SparkContext.setJobGroup]]
   * for more information.
   */
  def cancelJobGroup(groupId: String) {
    dagScheduler.cancelJobGroup(groupId)
  }

  /** Cancel all jobs that have been scheduled or are running.  */
  def cancelAllJobs() {
    dagScheduler.cancelAllJobs()
  }

  /**
   * Clean a closure to make it ready to serialized and send to tasks
   * (removes unreferenced variables in $outer's, updates REPL variables)
   */
  private[spark] def clean[F <: AnyRef](f: F): F = {
    ClosureCleaner.clean(f)
    f
  }

  /**
   * Set the directory under which RDDs are going to be checkpointed. The directory must
   * be a HDFS path if running on a cluster.
   */
  def setCheckpointDir(directory: String) {
    checkpointDir = Option(directory).map { dir =>
      val path = new Path(dir, UUID.randomUUID().toString)
      val fs = path.getFileSystem(hadoopConfiguration)
      fs.mkdirs(path)
      fs.getFileStatus(path).getPath.toString
    }
  }

  def getCheckpointDir = checkpointDir

  /** Default level of parallelism to use when not given by user (e.g. parallelize and makeRDD). */
  def defaultParallelism: Int = taskScheduler.defaultParallelism

  /** Default min number of partitions for Hadoop RDDs when not given by user */
  def defaultMinSplits: Int = math.min(defaultParallelism, 2)

  private val nextShuffleId = new AtomicInteger(0)

  private[spark] def newShuffleId(): Int = nextShuffleId.getAndIncrement()

  private val nextRddId = new AtomicInteger(0)

  /** Register a new RDD, returning its RDD ID */
  private[spark] def newRddId(): Int = nextRddId.getAndIncrement()

  /** Called by MetadataCleaner to clean up the persistentRdds map periodically */
  private[spark] def cleanup(cleanupTime: Long) {
    persistentRdds.clearOldValues(cleanupTime)
  }
}

/**
 * The SparkContext object contains a number of implicit conversions and parameters for use with
 * various Spark features.
 */
object SparkContext {

  private[spark] val SPARK_JOB_DESCRIPTION = "spark.job.description"

  private[spark] val SPARK_JOB_GROUP_ID = "spark.jobGroup.id"

  private[spark] val SPARK_UNKNOWN_USER = "<unknown>"

  implicit object DoubleAccumulatorParam extends AccumulatorParam[Double] {
    def addInPlace(t1: Double, t2: Double): Double = t1 + t2
    def zero(initialValue: Double) = 0.0
  }

  implicit object IntAccumulatorParam extends AccumulatorParam[Int] {
    def addInPlace(t1: Int, t2: Int): Int = t1 + t2
    def zero(initialValue: Int) = 0
  }

  implicit object LongAccumulatorParam extends AccumulatorParam[Long] {
    def addInPlace(t1: Long, t2: Long) = t1 + t2
    def zero(initialValue: Long) = 0L
  }

  implicit object FloatAccumulatorParam extends AccumulatorParam[Float] {
    def addInPlace(t1: Float, t2: Float) = t1 + t2
    def zero(initialValue: Float) = 0f
  }

  // TODO: Add AccumulatorParams for other types, e.g. lists and strings

  implicit def rddToPairRDDFunctions[K: ClassTag, V: ClassTag](rdd: RDD[(K, V)]) =
    new PairRDDFunctions(rdd)

  implicit def rddToAsyncRDDActions[T: ClassTag](rdd: RDD[T]) = new AsyncRDDActions(rdd)

  implicit def rddToSequenceFileRDDFunctions[K <% Writable: ClassTag, V <% Writable: ClassTag](
      rdd: RDD[(K, V)])   =
    new SequenceFileRDDFunctions(rdd)

  implicit def rddToOrderedRDDFunctions[K <% Ordered[K]: ClassTag, V: ClassTag](
      rdd: RDD[(K, V)]) =
    new OrderedRDDFunctions[K, V, (K, V)](rdd)

  implicit def doubleRDDToDoubleRDDFunctions(rdd: RDD[Double]) = new DoubleRDDFunctions(rdd)

  implicit def numericRDDToDoubleRDDFunctions[T](rdd: RDD[T])(implicit num: Numeric[T]) =
    new DoubleRDDFunctions(rdd.map(x => num.toDouble(x)))

  // Implicit conversions to common Writable types, for saveAsSequenceFile

  implicit def intToIntWritable(i: Int) = new IntWritable(i)

  implicit def longToLongWritable(l: Long) = new LongWritable(l)

  implicit def floatToFloatWritable(f: Float) = new FloatWritable(f)

  implicit def doubleToDoubleWritable(d: Double) = new DoubleWritable(d)

  implicit def boolToBoolWritable (b: Boolean) = new BooleanWritable(b)

  implicit def bytesToBytesWritable (aob: Array[Byte]) = new BytesWritable(aob)

  implicit def stringToText(s: String) = new Text(s)

  private implicit def arrayToArrayWritable[T <% Writable: ClassTag](arr: Traversable[T])
    : ArrayWritable = {
    def anyToWritable[U <% Writable](u: U): Writable = u

    new ArrayWritable(classTag[T].runtimeClass.asInstanceOf[Class[Writable]],
        arr.map(x => anyToWritable(x)).toArray)
  }

  // Helper objects for converting common types to Writable
  private def simpleWritableConverter[T, W <: Writable: ClassTag](convert: W => T) = {
    val wClass = classTag[W].runtimeClass.asInstanceOf[Class[W]]
    new WritableConverter[T](_ => wClass, x => convert(x.asInstanceOf[W]))
  }

  implicit def intWritableConverter() = simpleWritableConverter[Int, IntWritable](_.get)

  implicit def longWritableConverter() = simpleWritableConverter[Long, LongWritable](_.get)

  implicit def doubleWritableConverter() = simpleWritableConverter[Double, DoubleWritable](_.get)

  implicit def floatWritableConverter() = simpleWritableConverter[Float, FloatWritable](_.get)

  implicit def booleanWritableConverter() =
    simpleWritableConverter[Boolean, BooleanWritable](_.get)

  implicit def bytesWritableConverter() = {
    simpleWritableConverter[Array[Byte], BytesWritable](_.getBytes)
  }

  implicit def stringWritableConverter() = simpleWritableConverter[String, Text](_.toString)

  implicit def writableWritableConverter[T <: Writable]() =
    new WritableConverter[T](_.runtimeClass.asInstanceOf[Class[T]], _.asInstanceOf[T])

  /**
   * Find the JAR from which a given class was loaded, to make it easy for users to pass
   * their JARs to SparkContext.
   */
  def jarOfClass(cls: Class[_]): Seq[String] = {
    val uri = cls.getResource("/" + cls.getName.replace('.', '/') + ".class")
    if (uri != null) {
      val uriStr = uri.toString
      if (uriStr.startsWith("jar:file:")) {
        // URI will be of the form "jar:file:/path/foo.jar!/package/cls.class",
        // so pull out the /path/foo.jar
        List(uriStr.substring("jar:file:".length, uriStr.indexOf('!')))
      } else {
        Nil
      }
    } else {
      Nil
    }
  }

  /**
   * Find the JAR that contains the class of a particular object, to make it easy for users
   * to pass their JARs to SparkContext. In most cases you can call jarOfObject(this) in
   * your driver program.
   */
  def jarOfObject(obj: AnyRef): Seq[String] = jarOfClass(obj.getClass)

  /**
   * Creates a modified version of a SparkConf with the parameters that can be passed separately
   * to SparkContext, to make it easier to write SparkContext's constructors. This ignores
   * parameters that are passed as the default value of null, instead of throwing an exception
   * like SparkConf would.
   */
  private[spark] def updatedConf(
      conf: SparkConf,
      master: String,
      appName: String,
      sparkHome: String = null,
      jars: Seq[String] = Nil,
      environment: Map[String, String] = Map()): SparkConf =
  {
    val res = conf.clone()
    res.setMaster(master)
    res.setAppName(appName)
    if (sparkHome != null) {
      res.setSparkHome(sparkHome)
    }
    if (jars != null && !jars.isEmpty) {
      res.setJars(jars)
    }
    res.setExecutorEnv(environment.toSeq)
    res
  }

  /** Creates a task scheduler based on a given master URL. Extracted for testing. */
  private def createTaskScheduler(sc: SparkContext, master: String, appName: String)
      : TaskScheduler =
  {
    // Regular expression used for local[N] master format
    val LOCAL_N_REGEX = """local\[([0-9]+)\]""".r
    // Regular expression for local[N, maxRetries], used in tests with failing tasks
    val LOCAL_N_FAILURES_REGEX = """local\[([0-9]+)\s*,\s*([0-9]+)\]""".r
    // Regular expression for simulating a Spark cluster of [N, cores, memory] locally
    val LOCAL_CLUSTER_REGEX = """local-cluster\[\s*([0-9]+)\s*,\s*([0-9]+)\s*,\s*([0-9]+)\s*]""".r
    // Regular expression for connecting to Spark deploy clusters
    val SPARK_REGEX = """spark://(.*)""".r
    // Regular expression for connection to Mesos cluster by mesos:// or zk:// url
    val MESOS_REGEX = """(mesos|zk)://.*""".r
    // Regular expression for connection to Simr cluster
    val SIMR_REGEX = """simr://(.*)""".r

    // When running locally, don't try to re-execute tasks on failure.
    val MAX_LOCAL_TASK_FAILURES = 1

    master match {
      case "local" =>
        val scheduler = new TaskSchedulerImpl(sc, MAX_LOCAL_TASK_FAILURES, isLocal = true)
        val backend = new LocalBackend(scheduler, 1)
        scheduler.initialize(backend)
        scheduler

      case LOCAL_N_REGEX(threads) =>
        val scheduler = new TaskSchedulerImpl(sc, MAX_LOCAL_TASK_FAILURES, isLocal = true)
        val backend = new LocalBackend(scheduler, threads.toInt)
        scheduler.initialize(backend)
        scheduler

      case LOCAL_N_FAILURES_REGEX(threads, maxFailures) =>
        val scheduler = new TaskSchedulerImpl(sc, maxFailures.toInt, isLocal = true)
        val backend = new LocalBackend(scheduler, threads.toInt)
        scheduler.initialize(backend)
        scheduler

      case SPARK_REGEX(sparkUrl) =>
        val scheduler = new TaskSchedulerImpl(sc)
        val masterUrls = sparkUrl.split(",").map("spark://" + _)
        val backend = new SparkDeploySchedulerBackend(scheduler, sc, masterUrls, appName)
        scheduler.initialize(backend)
        scheduler

      case LOCAL_CLUSTER_REGEX(numSlaves, coresPerSlave, memoryPerSlave) =>
        // Check to make sure memory requested <= memoryPerSlave. Otherwise Spark will just hang.
        val memoryPerSlaveInt = memoryPerSlave.toInt
        if (sc.executorMemory > memoryPerSlaveInt) {
          throw new SparkException(
            "Asked to launch cluster with %d MB RAM / worker but requested %d MB/worker".format(
              memoryPerSlaveInt, sc.executorMemory))
        }

        val scheduler = new TaskSchedulerImpl(sc)
        val localCluster = new LocalSparkCluster(
          numSlaves.toInt, coresPerSlave.toInt, memoryPerSlaveInt)
        val masterUrls = localCluster.start()
        val backend = new SparkDeploySchedulerBackend(scheduler, sc, masterUrls, appName)
        scheduler.initialize(backend)
        backend.shutdownCallback = (backend: SparkDeploySchedulerBackend) => {
          localCluster.stop()
        }
        scheduler

      case "yarn-standalone" =>
        val scheduler = try {
          val clazz = Class.forName("org.apache.spark.scheduler.cluster.YarnClusterScheduler")
          val cons = clazz.getConstructor(classOf[SparkContext])
          cons.newInstance(sc).asInstanceOf[TaskSchedulerImpl]
        } catch {
          // TODO: Enumerate the exact reasons why it can fail
          // But irrespective of it, it means we cannot proceed !
          case th: Throwable => {
            throw new SparkException("YARN mode not available ?", th)
          }
        }
        val backend = new CoarseGrainedSchedulerBackend(scheduler, sc.env.actorSystem)
        scheduler.initialize(backend)
        scheduler

      case "yarn-client" =>
        val scheduler = try {
          val clazz =
            Class.forName("org.apache.spark.scheduler.cluster.YarnClientClusterScheduler")
          val cons = clazz.getConstructor(classOf[SparkContext])
          cons.newInstance(sc).asInstanceOf[TaskSchedulerImpl]

        } catch {
          case th: Throwable => {
            throw new SparkException("YARN mode not available ?", th)
          }
        }

        val backend = try {
          val clazz =
            Class.forName("org.apache.spark.scheduler.cluster.YarnClientSchedulerBackend")
          val cons = clazz.getConstructor(classOf[TaskSchedulerImpl], classOf[SparkContext])
          cons.newInstance(scheduler, sc).asInstanceOf[CoarseGrainedSchedulerBackend]
        } catch {
          case th: Throwable => {
            throw new SparkException("YARN mode not available ?", th)
          }
        }

        scheduler.initialize(backend)
        scheduler

      case mesosUrl @ MESOS_REGEX(_) =>
        MesosNativeLibrary.load()
        val scheduler = new TaskSchedulerImpl(sc)
        val coarseGrained = sc.conf.getBoolean("spark.mesos.coarse", false)
        val url = mesosUrl.stripPrefix("mesos://") // strip scheme from raw Mesos URLs
        val backend = if (coarseGrained) {
          new CoarseMesosSchedulerBackend(scheduler, sc, url, appName)
        } else {
          new MesosSchedulerBackend(scheduler, sc, url, appName)
        }
        scheduler.initialize(backend)
        scheduler

      case SIMR_REGEX(simrUrl) =>
        val scheduler = new TaskSchedulerImpl(sc)
        val backend = new SimrSchedulerBackend(scheduler, sc, simrUrl)
        scheduler.initialize(backend)
        scheduler

      case _ =>
        throw new SparkException("Could not parse Master URL: '" + master + "'")
    }
  }
}

/**
 * A class encapsulating how to convert some type T to Writable. It stores both the Writable class
 * corresponding to T (e.g. IntWritable for Int) and a function for doing the conversion.
 * The getter for the writable class takes a ClassTag[T] in case this is a generic object
 * that doesn't know the type of T when it is created. This sounds strange but is necessary to
 * support converting subclasses of Writable to themselves (writableWritableConverter).
 */
private[spark] class WritableConverter[T](
    val writableClass: ClassTag[T] => Class[_ <: Writable],
    val convert: Writable => T)
  extends Serializable