Criar cluster de escalonamento automático

Cria um cluster do Dataproc com uma política de escalonamento automático.

Exemplo de código

Java

Antes de testar este exemplo, siga as instruções de configuração do Java no Guia de início rápido do Managed Service for Apache Spark: como usar bibliotecas de cliente. Para mais informações, consulte a documentação de referência da API Java do Managed Service for Apache Spark.

Para autenticar o Managed Service for Apache Spark, configure o Application Default Credentials. Se quiser mais informações, consulte Configurar a autenticação para um ambiente de desenvolvimento local.

import com.google.api.gax.longrunning.OperationFuture;
import com.google.cloud.dataproc.v1.AutoscalingConfig;
import com.google.cloud.dataproc.v1.AutoscalingPolicy;
import com.google.cloud.dataproc.v1.AutoscalingPolicyServiceClient;
import com.google.cloud.dataproc.v1.AutoscalingPolicyServiceSettings;
import com.google.cloud.dataproc.v1.BasicAutoscalingAlgorithm;
import com.google.cloud.dataproc.v1.BasicYarnAutoscalingConfig;
import com.google.cloud.dataproc.v1.Cluster;
import com.google.cloud.dataproc.v1.ClusterConfig;
import com.google.cloud.dataproc.v1.ClusterControllerClient;
import com.google.cloud.dataproc.v1.ClusterControllerSettings;
import com.google.cloud.dataproc.v1.ClusterOperationMetadata;
import com.google.cloud.dataproc.v1.InstanceGroupAutoscalingPolicyConfig;
import com.google.cloud.dataproc.v1.InstanceGroupConfig;
import com.google.cloud.dataproc.v1.RegionName;
import com.google.protobuf.Duration;
import java.io.IOException;
import java.util.concurrent.ExecutionException;

public class CreateClusterWithAutoscaling {

  public static void createClusterwithAutoscaling() throws IOException, InterruptedException {
    // TODO(developer): Replace these variables before running the sample.
    String projectId = "your-project-id";
    String region = "your-project-region";
    String clusterName = "your-cluster-name";
    String autoscalingPolicyName = "your-autoscaling-policy";
    createClusterwithAutoscaling(projectId, region, clusterName, autoscalingPolicyName);
  }

  public static void createClusterwithAutoscaling(
      String projectId, String region, String clusterName, String autoscalingPolicyName)
      throws IOException, InterruptedException {
    String myEndpoint = String.format("%s-dataproc.googleapis.com:443", region);

    // Configure the settings for the cluster controller client.
    ClusterControllerSettings clusterControllerSettings =
        ClusterControllerSettings.newBuilder().setEndpoint(myEndpoint).build();

    // Configure the settings for the autoscaling policy service client.
    AutoscalingPolicyServiceSettings autoscalingPolicyServiceSettings =
        AutoscalingPolicyServiceSettings.newBuilder().setEndpoint(myEndpoint).build();

    // Create a cluster controller client and an autoscaling controller client with the configured
    // settings. The clients only need to be created once and can be reused for multiple requests.
    // Using a
    // try-with-resources closes the client, but this can also be done manually with the .close()
    // method.
    try (ClusterControllerClient clusterControllerClient =
            ClusterControllerClient.create(clusterControllerSettings);
        AutoscalingPolicyServiceClient autoscalingPolicyServiceClient =
            AutoscalingPolicyServiceClient.create(autoscalingPolicyServiceSettings)) {

      // Create the Autoscaling policy.
      InstanceGroupAutoscalingPolicyConfig workerInstanceGroupAutoscalingPolicyConfig =
          InstanceGroupAutoscalingPolicyConfig.newBuilder()
              .setMinInstances(2)
              .setMaxInstances(100)
              .setWeight(1)
              .build();
      InstanceGroupAutoscalingPolicyConfig secondaryWorkerInstanceGroupAutoscalingPolicyConfig =
          InstanceGroupAutoscalingPolicyConfig.newBuilder()
              .setMinInstances(0)
              .setMaxInstances(100)
              .setWeight(1)
              .build();
      BasicYarnAutoscalingConfig basicYarnApplicationConfig =
          BasicYarnAutoscalingConfig.newBuilder()
              .setScaleUpFactor(0.05)
              .setScaleDownFactor(1.0)
              .setScaleUpMinWorkerFraction(0.0)
              .setScaleUpMinWorkerFraction(0.0)
              .setGracefulDecommissionTimeout(Duration.newBuilder().setSeconds(3600).build())
              .build();
      BasicAutoscalingAlgorithm basicAutoscalingAlgorithm =
          BasicAutoscalingAlgorithm.newBuilder()
              .setCooldownPeriod(Duration.newBuilder().setSeconds(240).build())
              .setYarnConfig(basicYarnApplicationConfig)
              .build();
      AutoscalingPolicy autoscalingPolicy =
          AutoscalingPolicy.newBuilder()
              .setId(autoscalingPolicyName)
              .setWorkerConfig(workerInstanceGroupAutoscalingPolicyConfig)
              .setSecondaryWorkerConfig(secondaryWorkerInstanceGroupAutoscalingPolicyConfig)
              .setBasicAlgorithm(basicAutoscalingAlgorithm)
              .build();
      RegionName parent = RegionName.of(projectId, region);

      // Policy is uploaded here.
      autoscalingPolicyServiceClient.createAutoscalingPolicy(parent, autoscalingPolicy);

      // Now the policy can be referenced when creating a cluster.
      String autoscalingPolicyUri =
          String.format(
              "projects/%s/locations/%s/autoscalingPolicies/%s",
              projectId, region, autoscalingPolicyName);
      AutoscalingConfig autoscalingConfig =
          AutoscalingConfig.newBuilder().setPolicyUri(autoscalingPolicyUri).build();

      // Configure the settings for our cluster.
      InstanceGroupConfig masterConfig =
          InstanceGroupConfig.newBuilder()
              .setMachineTypeUri("n1-standard-2")
              .setNumInstances(1)
              .build();
      InstanceGroupConfig workerConfig =
          InstanceGroupConfig.newBuilder()
              .setMachineTypeUri("n1-standard-2")
              .setNumInstances(2)
              .build();
      ClusterConfig clusterConfig =
          ClusterConfig.newBuilder()
              .setMasterConfig(masterConfig)
              .setWorkerConfig(workerConfig)
              .setAutoscalingConfig(autoscalingConfig)
              .build();

      // Create the cluster object with the desired cluster config.
      Cluster cluster =
          Cluster.newBuilder().setClusterName(clusterName).setConfig(clusterConfig).build();

      // Create the Dataproc cluster.
      OperationFuture<Cluster, ClusterOperationMetadata> createClusterAsyncRequest =
          clusterControllerClient.createClusterAsync(projectId, region, cluster);
      Cluster response = createClusterAsyncRequest.get();

      // Print out a success message.
      System.out.printf("Cluster created successfully: %s", response.getClusterName());

    } catch (ExecutionException e) {
      // If cluster creation does not complete successfully, print the error message.
      System.err.println(String.format("createClusterWithAutoscaling: %s ", e.getMessage()));
    }
  }
}

A seguir

Para pesquisar e filtrar exemplos de código de outros Google Cloud produtos, consulte a Google Cloud pesquisa de exemplos de código.