Example usage for com.amazonaws.services.elasticmapreduce.model JobFlowInstancesConfig withEc2KeyName

Introduction

In this page you can find the example usage for com.amazonaws.services.elasticmapreduce.model JobFlowInstancesConfig withEc2KeyName.

Prototype

public JobFlowInstancesConfig withEc2KeyName(String ec2KeyName) 

Source Link

Document

The name of the EC2 key pair that can be used to ssh to the master node as the user called "hadoop."

Usage

From source file:awswc.AwsConsoleApp.java

License:Open Source License

static void runJobFlow() throws InterruptedException {
    // Configure instances to use
    JobFlowInstancesConfig instances = new JobFlowInstancesConfig();
    //********************************************************************//
    instances.setHadoopVersion(HADOOP_VERSION);
    instances.withEc2KeyName("ayuda-vp1");
    instances.setInstanceCount(MASTER_INSTANCE_COUNT);
    //instances.setInstanceGroups(instanceGroups)
    instances.setMasterInstanceType(InstanceType.M24xlarge.toString());
    instances.setSlaveInstanceType(InstanceType.M24xlarge.toString());
    //********************************************************************//
    HadoopJarStepConfig hadoopJarStep1 = new HadoopJarStepConfig().withJar(S3N_WORD_COUNT_JAR_) // This should be a full map reduce application.
            .withArgs(BUCKET_NAME + "inWC", BUCKET_NAME + "outWC");

    StepConfig stepConfig1 = new StepConfig().withName("wordcount").withHadoopJarStep(hadoopJarStep1)
            .withActionOnFailure("TERMINATE_JOB_FLOW");

    //********************************************************************//

    //********************************************************************//
    HadoopJarStepConfig hadoopJarStep2 = new HadoopJarStepConfig().withJar(S3N_MAX_WORD_COUNT_JAR) // This should be a full map reduce application.
            .withArgs(BUCKET_NAME + "outWC", BUCKET_NAME + "outXWC", "hate", "10");

    StepConfig stepConfig2 = new StepConfig().withName("maxwordcount").withHadoopJarStep(hadoopJarStep2)
            .withActionOnFailure("TERMINATE_JOB_FLOW");
    //********************************************************************//

    Collection<StepConfig> csc = new ArrayList<StepConfig>();
    csc.add(stepConfig1);/* ww w  .  j a va 2  s. c o  m*/
    csc.add(stepConfig2);

    // BootstrapActions bootstrapActions = new BootstrapActions();
    RunJobFlowRequest runFlowRequest = new RunJobFlowRequest().withName(FLOW_NAME).withInstances(instances)
            .withSteps(csc).withLogUri(BUCKET_NAME + "debug")
    /*.withBootstrapActions(
      bootstrapActions.newRunIf(
          "instance.isMaster=true",
          bootstrapActions.newConfigureDaemons()
              .withHeapSize(Daemon.JobTracker, 4096)
              .build()),
              bootstrapActions.newRunIf(
                      "instance.isRunningNameNode=true",
                      bootstrapActions.newConfigureDaemons()
                      .withHeapSize(Daemon.NameNode, 4096).build()),
      bootstrapActions.newRunIf(
                                     "instance.isRunningDataNode=true",
                                     bootstrapActions.newConfigureDaemons()
                                     .withHeapSize(Daemon.DataNode, 4096).build()),
      bootstrapActions.newRunIf(
                                     "instance.isRunningJobTracker=true",
                                     bootstrapActions.newConfigureDaemons()
                                     .withHeapSize(Daemon.JobTracker, 4096).build()),
      bootstrapActions.newRunIf(
                                             "instance.isRunningTaskTracker=true",
                                             bootstrapActions.newConfigureDaemons()
                                             .withHeapSize(Daemon.TaskTracker, 4096).build())                                             
                                             
                                     /*,
                                             
      bootstrapActions.newRunIf(
                                             "instance.isSlave=true",
      bootstrapActions.newConfigureHadoop()
                                       .withKeyValue(ConfigFile.Site,"mapred.tasktracker.map.tasks.maximum", "4"))                                            
              )*/;

    RunJobFlowResult runJobFlowResult = emr.runJobFlow(runFlowRequest);

    String jobFlowId = runJobFlowResult.getJobFlowId();
    System.out.println("Ran job flow with id: " + jobFlowId);

    //wasFinished(runJobFlowResult);

}

---