List of usage examples for com.amazonaws.services.elasticmapreduce.model JobFlowInstancesConfig setHadoopVersion
public void setHadoopVersion(String hadoopVersion)
Applies only to Amazon EMR release versions earlier than 4.0.
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);// w ww . j av a 2 s.com 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); }
From source file:com.clouddrive.parth.NewClass.java
private static JobFlowInstancesConfig configInstance() throws Exception { // Configure instances to use JobFlowInstancesConfig instance = new JobFlowInstancesConfig(); instance.setHadoopVersion(HADOOP_VERSION); instance.setInstanceCount(INSTANCE_COUNT); instance.setMasterInstanceType(INSTANCE_TYPE); instance.setSlaveInstanceType(INSTANCE_TYPE); //instance.setKeepJobFlowAliveWhenNoSteps(true); //instance.setEc2KeyName("parth"); return instance; }
From source file:datameer.awstasks.aws.emr.EmrCluster.java
License:Apache License
public synchronized void startup() throws InterruptedException { checkConnection(false);/*from w ww.j a v a2 s .c o m*/ _clusterState = ClusterState.STARTING; boolean successful = false; try { EmrSettings settings = getSettings(); if (settings.getPrivateKeyName() == null) { throw new NullPointerException( "privateKeyName must not be null please configure settings properly"); } LOG.info("Starting job flow '" + getName() + "' ..."); if (!getRunningJobFlowDetailsByName(getName()).isEmpty()) { throw new IllegalStateException("Job flow with name '" + getName() + "' already running."); } boolean keepAlive = true; JobFlowInstancesConfig jobConfig = new JobFlowInstancesConfig(); jobConfig.setHadoopVersion(_settings.getHadoopVersion()); jobConfig.setMasterInstanceType(settings.getMasterInstanceType().getId()); jobConfig.setSlaveInstanceType(settings.getNodeInstanceType().getId()); jobConfig.setInstanceCount(settings.getInstanceCount()); jobConfig.setEc2KeyName(settings.getPrivateKeyName()); jobConfig.setPlacement(new PlacementType()); jobConfig.setKeepJobFlowAliveWhenNoSteps(keepAlive); final RunJobFlowRequest startRequest = new RunJobFlowRequest(); startRequest.setLogUri("s3n://" + settings.getS3Bucket() + settings.getS3LogPath()); startRequest.setInstances(jobConfig); startRequest.setName(getName()); startRequest.setAdditionalInfo(_settings.getAdditionalStartInfo()); startRequest.setBootstrapActions(_settings.getBootstrapActions()); if (settings.isDebugEnabled()) { startRequest.withSteps(DEBUG_STEP); } RunJobFlowResult startResponse = _emrWebService.runJobFlow(startRequest); _jobFlowId = startResponse.getJobFlowId(); waitUntilClusterStarted(_jobFlowId); LOG.info("elastic cluster '" + getName() + "/" + _jobFlowId + "' started, master-host is " + _masterHost); successful = true; } finally { if (successful) { _clusterState = ClusterState.CONNECTED; } else { _clusterState = ClusterState.UNCONNECTED; _jobFlowId = null; } } }
From source file:org.finra.dm.dao.impl.EmrDaoImpl.java
License:Apache License
/** * Create the job flow instance configuration which contains all the job flow configuration details. * * @param emrClusterDefinition the EMR cluster definition that contains all the EMR parameters. * * @return the job flow instance configuration. *///www.j a va 2 s.c o m private JobFlowInstancesConfig getJobFlowInstancesConfig(EmrClusterDefinition emrClusterDefinition) { // Create a new job flow instance config object JobFlowInstancesConfig jobFlowInstancesConfig = new JobFlowInstancesConfig(); // Add the DM EMR support security group as additional group to master node. String additionalSecurityGroup = configurationHelper .getProperty(ConfigurationValue.EMR_DM_SUPPORT_SECURITY_GROUP); if (StringUtils.isNotBlank(additionalSecurityGroup)) { List<String> additionalSecurityGroups = new ArrayList<>(); additionalSecurityGroups.add(additionalSecurityGroup); jobFlowInstancesConfig.setAdditionalMasterSecurityGroups(additionalSecurityGroups); } // Fill-in the ssh key if (StringUtils.isNotBlank(emrClusterDefinition.getSshKeyPairName())) { jobFlowInstancesConfig.setEc2KeyName(emrClusterDefinition.getSshKeyPairName()); } // Fill-in subnet id if (StringUtils.isNotBlank(emrClusterDefinition.getSubnetId())) { jobFlowInstancesConfig.setEc2SubnetId(emrClusterDefinition.getSubnetId()); } // Fill in instance groups jobFlowInstancesConfig.setInstanceGroups(getInstanceGroupConfig(emrClusterDefinition)); // Check for optional parameters and then fill-in // Keep Alive Cluster flag if (emrClusterDefinition.isKeepAlive() != null) { jobFlowInstancesConfig.setKeepJobFlowAliveWhenNoSteps(emrClusterDefinition.isKeepAlive()); } // Termination protection flag if (emrClusterDefinition.isTerminationProtection() != null) { jobFlowInstancesConfig.setTerminationProtected(emrClusterDefinition.isTerminationProtection()); } // Setting the hadoop version if (StringUtils.isNotBlank(emrClusterDefinition.getHadoopVersion())) { jobFlowInstancesConfig.setHadoopVersion(emrClusterDefinition.getHadoopVersion()); } // Return the object return jobFlowInstancesConfig; }
From source file:org.finra.herd.dao.impl.EmrDaoImpl.java
License:Apache License
/** * Creates the job flow instance configuration containing specification of the number and type of Amazon EC2 instances. * * @param emrClusterDefinition the EMR cluster definition that contains all the EMR parameters * * @return the job flow instance configuration *//*from w ww . j a va 2s. c o m*/ private JobFlowInstancesConfig getJobFlowInstancesConfig(EmrClusterDefinition emrClusterDefinition) { // Create a new job flow instances configuration object. JobFlowInstancesConfig jobFlowInstancesConfig = new JobFlowInstancesConfig(); // Set up master/slave security group jobFlowInstancesConfig.setEmrManagedMasterSecurityGroup(emrClusterDefinition.getMasterSecurityGroup()); jobFlowInstancesConfig.setEmrManagedSlaveSecurityGroup(emrClusterDefinition.getSlaveSecurityGroup()); // Add additional security groups to master nodes. jobFlowInstancesConfig .setAdditionalMasterSecurityGroups(emrClusterDefinition.getAdditionalMasterSecurityGroups()); // Add additional security groups to slave nodes. jobFlowInstancesConfig .setAdditionalSlaveSecurityGroups(emrClusterDefinition.getAdditionalSlaveSecurityGroups()); // Fill-in the ssh key. if (StringUtils.isNotBlank(emrClusterDefinition.getSshKeyPairName())) { jobFlowInstancesConfig.setEc2KeyName(emrClusterDefinition.getSshKeyPairName()); } // Fill in configuration for the instance groups in a cluster. jobFlowInstancesConfig .setInstanceGroups(getInstanceGroupConfigs(emrClusterDefinition.getInstanceDefinitions())); // Fill in instance fleet configuration. jobFlowInstancesConfig.setInstanceFleets(getInstanceFleets(emrClusterDefinition.getInstanceFleets())); // Fill-in subnet id. if (StringUtils.isNotBlank(emrClusterDefinition.getSubnetId())) { // Use collection of subnet IDs when instance fleet configuration is specified. Otherwise, we expect a single EC2 subnet ID to be passed here. if (CollectionUtils.isNotEmpty(jobFlowInstancesConfig.getInstanceFleets())) { jobFlowInstancesConfig .setEc2SubnetIds(herdStringHelper.splitAndTrim(emrClusterDefinition.getSubnetId(), ",")); } else { jobFlowInstancesConfig.setEc2SubnetId(emrClusterDefinition.getSubnetId()); } } // Fill in optional keep alive flag. if (emrClusterDefinition.isKeepAlive() != null) { jobFlowInstancesConfig.setKeepJobFlowAliveWhenNoSteps(emrClusterDefinition.isKeepAlive()); } // Fill in optional termination protection flag. if (emrClusterDefinition.isTerminationProtection() != null) { jobFlowInstancesConfig.setTerminationProtected(emrClusterDefinition.isTerminationProtection()); } // Fill in optional Hadoop version flag. if (StringUtils.isNotBlank(emrClusterDefinition.getHadoopVersion())) { jobFlowInstancesConfig.setHadoopVersion(emrClusterDefinition.getHadoopVersion()); } // Return the object. return jobFlowInstancesConfig; }
From source file:org.huahinframework.emanager.amazonaws.elasticmapreduce.ElasticMapReduceManager.java
License:Apache License
/** * @return JobFlowInstancesConfig//w ww . j ava 2 s .c o m */ private JobFlowInstancesConfig setupJobFlowInstancesConfig() { JobFlowInstancesConfig config = new JobFlowInstancesConfig().withKeepJobFlowAliveWhenNoSteps(true) .withInstanceCount(emrProperties.getInstanceCount()) .withMasterInstanceType(emrProperties.getMasterInstanceType()); if (!isEmpty(emrProperties.getKeyPairName())) { config.setEc2KeyName(emrProperties.getKeyPairName()); } if (!isEmpty(emrProperties.getHadoopVersion())) { config.setHadoopVersion(emrProperties.getHadoopVersion()); } if (!isEmpty(emrProperties.getAvailabilityZone())) { config.setPlacement(new PlacementType().withAvailabilityZone(emrProperties.getAvailabilityZone())); } if (!isEmpty(emrProperties.getSlaveInstanceType())) { config.setSlaveInstanceType(emrProperties.getSlaveInstanceType()); } else { config.setSlaveInstanceType(emrProperties.getMasterInstanceType()); } return config; }
From source file:org.pentaho.amazon.emr.job.AmazonElasticMapReduceJobExecutor.java
License:Apache License
public RunJobFlowRequest createJobFlow(String stagingS3BucketUrl, String stagingS3Jar, String mainClass) { List<String> jarStepArgs = new ArrayList<String>(); if (!StringUtil.isEmpty(cmdLineArgs)) { StringTokenizer st = new StringTokenizer(cmdLineArgs, " "); while (st.hasMoreTokens()) { String token = st.nextToken(); logBasic("adding args: " + token); jarStepArgs.add(token);/* w ww .ja va2 s. c o m*/ } } HadoopJarStepConfig hadoopJarStep = new HadoopJarStepConfig(); hadoopJarStep.setJar(stagingS3Jar); hadoopJarStep.setMainClass(mainClass); hadoopJarStep.setArgs(jarStepArgs); StepConfig stepConfig = new StepConfig(); stepConfig.setName("custom jar: " + jarUrl); stepConfig.setHadoopJarStep(hadoopJarStep); List<StepConfig> steps = new ArrayList<StepConfig>(); steps.add(stepConfig); String numInstancesS = environmentSubstitute(numInstances); int numInsts = 2; try { numInsts = Integer.parseInt(numInstancesS); } catch (NumberFormatException e) { logError("Unable to parse number of instances to use '" + numInstancesS + "' - " + "using 2 instances..."); } JobFlowInstancesConfig instances = new JobFlowInstancesConfig(); instances.setInstanceCount(numInsts); instances.setMasterInstanceType(getInstanceType(masterInstanceType)); instances.setSlaveInstanceType(getInstanceType(slaveInstanceType)); instances.setHadoopVersion("0.20"); RunJobFlowRequest runJobFlowRequest = new RunJobFlowRequest(); runJobFlowRequest.setSteps(steps); runJobFlowRequest.setLogUri(stagingS3BucketUrl); runJobFlowRequest.setName(hadoopJobName); runJobFlowRequest.setInstances(instances); // ScriptBootstrapActionConfig scriptBootstrapAction = new ScriptBootstrapActionConfig(); // scriptBootstrapAction.setPath("s3://mddwordcount/bootstrap.sh"); // List<String> bootstrapArgs = new ArrayList<String>(); // bootstrapArgs.add("http://pdi-node-dist.s3.amazonaws.com"); // // // bootstrapArgs.add( // "http://ci.pentaho.com/view/Data%20Integration/job/Kettle/lastSuccessfulBuild/artifact/Kettle/"); // bootstrapArgs.add("pdi-hadoop-node-TRUNK-SNAPSHOT.zip"); // scriptBootstrapAction.setArgs(bootstrapArgs); // BootstrapActionConfig bootstrapActionConfig = new BootstrapActionConfig(); // bootstrapActionConfig.setName("mdd bootstrap"); // bootstrapActionConfig.setScriptBootstrapAction(scriptBootstrapAction); // List<BootstrapActionConfig> bootstrapActions = new ArrayList<BootstrapActionConfig>(); // bootstrapActions.add(bootstrapActionConfig); // runJobFlowRequest.setBootstrapActions(bootstrapActions); return runJobFlowRequest; }
From source file:org.pentaho.amazon.hive.job.AmazonHiveJobExecutor.java
License:Apache License
/** * Prepare to create a EMR job flow./*from w w w . ja v a2 s .c om*/ * * @return RunJobFlowRequest The object to request an EMR job flow */ public RunJobFlowRequest createJobFlow() { // Create a RunJobFlowRequest object, set a name for the job flow. RunJobFlowRequest runJobFlowRequest = new RunJobFlowRequest(); runJobFlowRequest.setName(hadoopJobName); // Set a log URL. String logUrl = stagingDir; if (stagingDir.indexOf("@s3") > 0) { //$NON-NLS-1$ logUrl = S3FileProvider.SCHEME + "://" + stagingDir.substring(stagingDir.indexOf("@s3") + 4); //$NON-NLS-1$ } runJobFlowRequest.setLogUri(logUrl); // Determine the instances for Hadoop cluster. String numInstancesS = environmentSubstitute(numInstances); int numInsts = 2; try { numInsts = Integer.parseInt(numInstancesS); } catch (NumberFormatException e) { logError(BaseMessages.getString(PKG, "AmazonElasticMapReduceJobExecutor.InstanceNumber.Error", //$NON-NLS-1$ numInstancesS)); } JobFlowInstancesConfig instances = new JobFlowInstancesConfig(); instances.setInstanceCount(numInsts); instances.setMasterInstanceType(getInstanceType(masterInstanceType)); instances.setSlaveInstanceType(getInstanceType(slaveInstanceType)); instances.setHadoopVersion("0.20"); //$NON-NLS-1$ instances.setKeepJobFlowAliveWhenNoSteps(alive); runJobFlowRequest.setInstances(instances); // Set bootstrap actions. runJobFlowRequest.setBootstrapActions(ConfigBootstrapActions()); // Create an EMR step to setup Hive. String args = "s3://elasticmapreduce/libs/hive/hive-script --base-path s3://elasticmapreduce/libs/hive/ --hive-versions 0.7 --install-hive"; //$NON-NLS-1$ List<StepConfig> steps = ConfigHadoopJarStep("Setup Hive", jarUrl, args); //$NON-NLS-1$ runJobFlowRequest.setSteps(steps); return runJobFlowRequest; }