MapReduce之Job提交流程源码和切片源码分析

7次阅读

共计 10450 个字符，预计需要花费 27 分钟才能阅读完成。

hadoop2.7.2 MapReduce Job 提交源码及切片源码分析

  1. 首先从 waitForCompletion 函数进入
boolean result = job.waitForCompletion(true);
/**
   * Submit the job to the cluster and wait for it to finish.
   * @param verbose print the progress to the user
   * @return true if the job succeeded
   * @throws IOException thrown if the communication with the 
   *         <code>JobTracker</code> is lost
   */
  public boolean waitForCompletion(boolean verbose) throws IOException, InterruptedException,
                                            ClassNotFoundException {// 首先判断 state，当 state 为 DEFINE 时可以提交，进入 submit() 方法
    if (state == JobState.DEFINE) {submit();
    }
    if (verbose) {monitorAndPrintJob();
    } else {
      // get the completion poll interval from the client.
      int completionPollIntervalMillis = 
        Job.getCompletionPollInterval(cluster.getConf());
      while (!isComplete()) {
        try {Thread.sleep(completionPollIntervalMillis);
        } catch (InterruptedException ie) {}}
    }
    return isSuccessful();}
  1. 进入 submit() 方法
/**
   * Submit the job to the cluster and return immediately.
   * @throws IOException
   */
  public void submit() 
         throws IOException, InterruptedException, ClassNotFoundException {
    // 确认 JobState 状态为可提交状态，否则不能提交
    ensureState(JobState.DEFINE);
    // 设置使用最新的 API
    setUseNewAPI();
    // 进入 connect()方法，MapReduce 作业提交时连接集群是通过 Job 类的 connect()方法实现的，// 它实际上是构造集群 Cluster 实例 cluster
    connect();
    // connect()方法执行完之后，定义提交者 submitter
    final JobSubmitter submitter = 
        getJobSubmitter(cluster.getFileSystem(), cluster.getClient());
    status = ugi.doAs(new PrivilegedExceptionAction<JobStatus>() {public JobStatus run() throws IOException, InterruptedException, 
      ClassNotFoundException {// 这里的核心方法是 submitJobInternal(), 顾名思义，提交 job 的内部方法，实现了提交 job 的所有业务逻辑
          // 进入 submitJobInternal
        return submitter.submitJobInternal(Job.this, cluster);
      }
    });
    // 提交之后 state 状态改变
    state = JobState.RUNNING;
    LOG.info("The url to track the job:" + getTrackingURL());
   }
  1. 进入 connect() 方法
  • MapReduce 作业提交时连接集群通过 Job 的 Connect 方法实现，它实际上是构造集群 Cluster 实例 cluster
  • cluster 是连接 MapReduce 集群的一种工具，提供了获取 MapReduce 集群信息的方法
  • 在 Cluster 内部，有一个与集群进行通信的客户端通信协议 ClientProtocol 的实例 client，它由 ClientProtocolProvider 的静态 create()方法构造
  • 在 create 内部，Hadoop2.x 中提供了两种模式的 ClientProtocol，分别为 Yarn 模式的 YARNRunner 和 Local 模式的 LocalJobRunner，Cluster 实际上是由它们负责与集群进行通信的
  private synchronized void connect()
          throws IOException, InterruptedException, ClassNotFoundException {if (cluster == null) {// cluster 提供了远程获取 MapReduce 的方法
      cluster = 
        ugi.doAs(new PrivilegedExceptionAction<Cluster>() {public Cluster run()
                          throws IOException, InterruptedException, 
                                 ClassNotFoundException {// 只需关注这个 Cluster()构造器，构造集群 cluster 实例
                     return new Cluster(getConfiguration());
                   }
                 });
    }
  }
  1. 进入 Cluster() 构造器
// 首先调用一个参数的构造器，间接调用两个参数的构造器
public Cluster(Configuration conf) throws IOException {this(null, conf);
  }

  public Cluster(InetSocketAddress jobTrackAddr, Configuration conf) 
      throws IOException {
    this.conf = conf;
    this.ugi = UserGroupInformation.getCurrentUser();
    // 最重要的 initialize 方法
    initialize(jobTrackAddr, conf);
  }
  
// cluster 中要关注的两个成员变量是客户端通讯协议提供者 ClientProtocolProvider 和客户端通讯协议 ClientProtocol 实例 client
  private void initialize(InetSocketAddress jobTrackAddr, Configuration conf)
      throws IOException {synchronized (frameworkLoader) {for (ClientProtocolProvider provider : frameworkLoader) {
        LOG.debug("Trying ClientProtocolProvider :"
            + provider.getClass().getName());
        ClientProtocol clientProtocol = null; 
        try {
          // 如果配置文件没有配置 YARN 信息，则构建 LocalRunner,MR 任务本地运行
          // 如果配置文件有配置 YARN 信息，则构建 YarnRunner,MR 任务在 YARN 集群上运行
          if (jobTrackAddr == null) {// 客户端通讯协议 client 是调用 ClientProtocolProvider 的 create()方法实现
            clientProtocol = provider.create(conf);
          } else {clientProtocol = provider.create(jobTrackAddr, conf);
          }

          if (clientProtocol != null) {
            clientProtocolProvider = provider;
            client = clientProtocol;
            LOG.debug("Picked" + provider.getClass().getName()
                + "as the ClientProtocolProvider");
            break;
          }
          else {LOG.debug("Cannot pick" + provider.getClass().getName()
                + "as the ClientProtocolProvider - returned null protocol");
          }
        } 
        catch (Exception e) {LOG.info("Failed to use" + provider.getClass().getName()
              + "due to error:", e);
        }
      }
    }

    if (null == clientProtocolProvider || null == client) {
      throw new IOException(
          "Cannot initialize Cluster. Please check your configuration for"
              + MRConfig.FRAMEWORK_NAME
              + "and the correspond server addresses.");
    }
  }
  1. 进入submitJobInternal()，job 的内部提交方法，用于提交 job 到集群
JobStatus submitJobInternal(Job job, Cluster cluster) 
  throws ClassNotFoundException, InterruptedException, IOException {

    //validate the jobs output specs 
    // 检查结果的输出路径是否已经存在，如果存在会报异常
    checkSpecs(job);

    // conf 里边是集群的 xml 配置文件信息
    Configuration conf = job.getConfiguration();
    // 添加 MR 框架到分布式缓存中
    addMRFrameworkToDistributedCache(conf);

    // 获取提交执行时相关资源的临时存放路径
    // 参数未配置时默认是（工作空间根目录下的）/tmp/hadoop-yarn/staging/ 提交作业用户名 /.staging
    Path jobStagingArea = JobSubmissionFiles.getStagingDir(cluster, conf);
    //configure the command line options correctly on the submitting dfs
    InetAddress ip = InetAddress.getLocalHost();
    if (ip != null) {// 记录提交作业的主机 IP、主机名，并且设置配置信息 conf
      submitHostAddress = ip.getHostAddress();
      submitHostName = ip.getHostName();
      conf.set(MRJobConfig.JOB_SUBMITHOST,submitHostName);
      conf.set(MRJobConfig.JOB_SUBMITHOSTADDR,submitHostAddress);
    }
    // 获取 JobId
    JobID jobId = submitClient.getNewJobID();
    // 设置 jobId
    job.setJobID(jobId);
    // 提交作业的路径 Path(Path parent, String child)，会将两个参数拼接为一个路径
    Path submitJobDir = new Path(jobStagingArea, jobId.toString());
    // job 的状态
    JobStatus status = null;
    try {
      conf.set(MRJobConfig.USER_NAME,
          UserGroupInformation.getCurrentUser().getShortUserName());
      conf.set("hadoop.http.filter.initializers", 
          "org.apache.hadoop.yarn.server.webproxy.amfilter.AmFilterInitializer");
      conf.set(MRJobConfig.MAPREDUCE_JOB_DIR, submitJobDir.toString());
      LOG.debug("Configuring job" + jobId + "with" + submitJobDir 
          + "as the submit dir");
      // get delegation token for the dir
      TokenCache.obtainTokensForNamenodes(job.getCredentials(),
          new Path[] { submitJobDir}, conf);
      
      populateTokenCache(conf, job.getCredentials());

      // generate a secret to authenticate shuffle transfers
      if (TokenCache.getShuffleSecretKey(job.getCredentials()) == null) {
        KeyGenerator keyGen;
        try {keyGen = KeyGenerator.getInstance(SHUFFLE_KEYGEN_ALGORITHM);
          keyGen.init(SHUFFLE_KEY_LENGTH);
        } catch (NoSuchAlgorithmException e) {throw new IOException("Error generating shuffle secret key", e);
        }
        SecretKey shuffleKey = keyGen.generateKey();
        TokenCache.setShuffleSecretKey(shuffleKey.getEncoded(),
            job.getCredentials());
      }
      if (CryptoUtils.isEncryptedSpillEnabled(conf)) {conf.setInt(MRJobConfig.MR_AM_MAX_ATTEMPTS, 1);
        LOG.warn("Max job attempts set to 1 since encrypted intermediate" +
                "data spill is enabled");
      }

      // 拷贝 jar 包到集群
      // 此方法中调用如下方法：rUploader.uploadFiles(job, jobSubmitDir);
      // uploadFiles 方法将 jar 包拷贝到集群
      copyAndConfigureFiles(job, submitJobDir);

      Path submitJobFile = JobSubmissionFiles.getJobConfPath(submitJobDir);
      
      // Create the splits for the job
      LOG.debug("Creating splits at" + jtFs.makeQualified(submitJobDir));
      // 计算切片，生成切片规划文件
      int maps = writeSplits(job, submitJobDir);
      conf.setInt(MRJobConfig.NUM_MAPS, maps);
      LOG.info("number of splits:" + maps);

      // write "queue admins of the queue to which job is being submitted"
      // to job file.
      String queue = conf.get(MRJobConfig.QUEUE_NAME,
          JobConf.DEFAULT_QUEUE_NAME);
      AccessControlList acl = submitClient.getQueueAdmins(queue);
      conf.set(toFullPropertyName(queue,
          QueueACL.ADMINISTER_JOBS.getAclName()), acl.getAclString());

      // removing jobtoken referrals before copying the jobconf to HDFS
      // as the tasks don't need this setting, actually they may break
      // because of it if present as the referral will point to a
      // different job.
      TokenCache.cleanUpTokenReferral(conf);

      if (conf.getBoolean(
          MRJobConfig.JOB_TOKEN_TRACKING_IDS_ENABLED,
          MRJobConfig.DEFAULT_JOB_TOKEN_TRACKING_IDS_ENABLED)) {
        // Add HDFS tracking ids
        ArrayList<String> trackingIds = new ArrayList<String>();
        for (Token<? extends TokenIdentifier> t :
            job.getCredentials().getAllTokens()) {trackingIds.add(t.decodeIdentifier().getTrackingId());
        }
        conf.setStrings(MRJobConfig.JOB_TOKEN_TRACKING_IDS,
            trackingIds.toArray(new String[trackingIds.size()]));
      }

      // Set reservation info if it exists
      ReservationId reservationId = job.getReservationId();
      if (reservationId != null) {conf.set(MRJobConfig.RESERVATION_ID, reservationId.toString());
      }

      // Write job file to submit dir
      writeConf(conf, submitJobFile);
      
      //
      // Now, actually submit the job (using the submit name)
      // 开始正式提交 job
      printTokens(jobId, job.getCredentials());
      status = submitClient.submitJob(jobId, submitJobDir.toString(), job.getCredentials());
      if (status != null) {return status;} else {throw new IOException("Could not launch job");
      }
    } finally {if (status == null) {LOG.info("Cleaning up the staging area" + submitJobDir);
        if (jtFs != null && submitJobDir != null)
          jtFs.delete(submitJobDir, true);

      }
    }
  }
  1. 进入writeSplits(job, submitJobDir)，计算切片，生成切片规划文件
  • 内部会调用 writeNewSplits(job, jobSubmitDir) 方法
  • writeNewSplits(job, jobSubmitDir)内部定义了一个 InputFormat 类型的实例 input

  • InputFormat 主要作用：

    • 验证 job 的输入规范
    • 对输入的文件进行切分，形成多个 InputSplit（切片）文件，每一个 InputSplit 对应着一个 map 任务（MapTask）
    • 将切片后的数据按照规则形成 key，value 键值对 RecordReader
  • input 调用 getSplits（）方法：List<InputSplit> splits = input.getSplits(job);
  1. 进入 FileInputFormat 类下的 getSplits(job) 方法
/** 
   * Generate the list of files and make them into FileSplits.
   * @param job the job context
   * @throws IOException
   */
  public List<InputSplit> getSplits(JobContext job) throws IOException {StopWatch sw = new StopWatch().start();
      
    // getFormatMinSplitSize()返回值固定为 1，getMinSplitSize(job)返回 job 大小
    long minSize = Math.max(getFormatMinSplitSize(), getMinSplitSize(job));
    // getMaxSplitSize(job)返回 Lang 类型的最大值
    long maxSize = getMaxSplitSize(job);

    // generate splits 生成切片
    List<InputSplit> splits = new ArrayList<InputSplit>();
    List<FileStatus> files = listStatus(job);
    // 遍历 job 下的所有文件
    for (FileStatus file: files) {
      // 获取文件路径
      Path path = file.getPath();
      // 获取文件大小
      long length = file.getLen();
      if (length != 0) {BlockLocation[] blkLocations;
        if (file instanceof LocatedFileStatus) {blkLocations = ((LocatedFileStatus) file).getBlockLocations();} else {FileSystem fs = path.getFileSystem(job.getConfiguration());
          blkLocations = fs.getFileBlockLocations(file, 0, length);
        }
        // 判断是否可分割
        if (isSplitable(job, path)) {
          // 获取块大小
          // 本地环境块大小默认为 32MB，YARN 环境在 hadoop2.x 新版本为 128MB，旧版本为 64MB
          long blockSize = file.getBlockSize();
          // 计算切片的逻辑大小，默认等于块大小
          // 返回值为：return Math.max(minSize, Math.min(maxSize, blockSize));
          // 其中 minSize=1，maxSize=Long 类型最大值，blockSize 为切片大小
          long splitSize = computeSplitSize(blockSize, minSize, maxSize);

          long bytesRemaining = length;
          // 每次切片时就要判断切片剩下的部分是否大于切片大小的 SPLIT_SLOP（默认为 1.1）倍，// 否则就不再切分，划为一块
          while (((double) bytesRemaining)/splitSize > SPLIT_SLOP) {int blkIndex = getBlockIndex(blkLocations, length-bytesRemaining);
            splits.add(makeSplit(path, length-bytesRemaining, splitSize,
                        blkLocations[blkIndex].getHosts(),
                        blkLocations[blkIndex].getCachedHosts()));
            bytesRemaining -= splitSize;
          }

          if (bytesRemaining != 0) {int blkIndex = getBlockIndex(blkLocations, length-bytesRemaining);
            splits.add(makeSplit(path, length-bytesRemaining, bytesRemaining,
                       blkLocations[blkIndex].getHosts(),
                       blkLocations[blkIndex].getCachedHosts()));
          }
        } else { // not splitable
          splits.add(makeSplit(path, 0, length, blkLocations[0].getHosts(),
                      blkLocations[0].getCachedHosts()));
        }
      } else { 
        //Create empty hosts array for zero length files
        splits.add(makeSplit(path, 0, length, new String[0]));
      }
    }
    // Save the number of input files for metrics/loadgen
    job.getConfiguration().setLong(NUM_INPUT_FILES, files.size());
    sw.stop();
    if (LOG.isDebugEnabled()) {LOG.debug("Total # of splits generated by getSplits:" + splits.size()
          + ", TimeTaken:" + sw.now(TimeUnit.MILLISECONDS));
    }
    return splits;
  }

正文完
发表至： hadoop
2019-09-29
 0
关于hadoop:hadoop教程hdfs客户端开发
关于hadoop:Hadoop-集群安装
2020版冲刺年薪30W超全大数据学习路线思维导图
关于hadoop:Hadoop之MapReduce四序列化和排序