「Managed Service for Apache Spark」は、以前は「Compute Engine 上の Dataproc」（クラスタデプロイ）と「Apache Spark 用 Google Cloud Serverless」（サーバーレスデプロイ）と呼ばれていたプロダクトの新しい名前です。

Google uses AI technology to translate content into your preferred language. AI translations can contain errors.

クライアントライブラリを使用してクラスタを作成する

下に記載のサンプルコードは、Cloud クライアントライブラリを使用して Managed Service for Apache Spark クラスタを作成し、そのクラスタでジョブを実行した後、そのクラスタを削除する方法を示しています。

以下を使用してこうしたタスクを実行することもできます。

API Explorer を使用したクイックスタートの API REST リクエスト
コンソールを使用してクラスタを作成するのコンソール Google Cloud Google Cloud
Google Cloud CLI を使用してクラスタを作成するの Google Cloud CLI

始める前に

アカウントにログインします。 Google Cloud を初めて使用する場合は、アカウントを作成して、実際のシナリオで Google プロダクトのパフォーマンスを評価してください。 Google Cloud新規のお客様には、ワークロードの実行、テスト、デプロイができる無料クレジット $300 分を差し上げます。

In the Google Cloud console, on the project selector page, select or create a Google Cloud project.

Roles required to select or create a project

Select a project: Selecting a project doesn't require a specific IAM role—you can select any project that you've been granted a role on.
Create a project: To create a project, you need the Project Creator role (roles/resourcemanager.projectCreator), which contains the resourcemanager.projects.create permission. Learn how to grant roles.

Go to project selector

Verify that billing is enabled for your Google Cloud project.

Enable the Dataproc API.

Roles required to enable APIs

To enable APIs, you need the Service Usage Admin IAM role (roles/serviceusage.serviceUsageAdmin), which contains the serviceusage.services.enable permission. Learn how to grant roles.

Enable the API

In the Google Cloud console, on the project selector page, select or create a Google Cloud project.

Roles required to select or create a project

Select a project: Selecting a project doesn't require a specific IAM role—you can select any project that you've been granted a role on.
Create a project: To create a project, you need the Project Creator role (roles/resourcemanager.projectCreator), which contains the resourcemanager.projects.create permission. Learn how to grant roles.

Go to project selector

Verify that billing is enabled for your Google Cloud project.

Enable the Dataproc API.

Roles required to enable APIs

To enable APIs, you need the Service Usage Admin IAM role (roles/serviceusage.serviceUsageAdmin), which contains the serviceusage.services.enable permission. Learn how to grant roles.

Enable the API

コードの実行

チュートリアルを試す: [Cloud Shell で開く] をクリックして、Python Cloud クライアントライブラリのチュートリアルを実行します。このチュートリアルでは、クラスタを作成し、PySpark ジョブを実行した後、クラスタを削除します。

Cloud Shell で開く

Go

クライアントライブラリをインストールします。詳しくは、開発環境の設定をご覧ください。
認証の設定

サンプル GitHub コードのクローンを作成して実行します。

PySpark 入力ファイルの選択: 下に記載の Google クライアントライブラリのコードサンプルでは、入力パラメータとして指定した PySpark ジョブを実行します。シンプルな「Hello World」PySpark アプリで渡すことが可能です。このアプリは Cloud Storage の gs://dataproc-examples/pyspark/hello-world/hello-world.py に配置されています。または独自の PySpark アプリで渡すことも可能です。Cloud Storage へのファイルのアップロードについての詳細は、オブジェクトのアップロードをご覧ください。


// This quickstart shows how you can use the Dataproc Client library to create a
// Dataproc cluster, submit a PySpark job to the cluster, wait for the job to finish
// and finally delete the cluster.
//
// Usage:
//
//	go build
//	./quickstart --project_id <PROJECT_ID> --region <REGION> \
//	    --cluster_name <CLUSTER_NAME> --job_file_path <GCS_JOB_FILE_PATH>
package main

import (
	"context"
	"flag"
	"fmt"
	"io"
	"log"
	"regexp"

	dataproc "cloud.google.com/go/dataproc/apiv1"
	"cloud.google.com/go/dataproc/apiv1/dataprocpb"
	"cloud.google.com/go/storage"
	"google.golang.org/api/option"
)

func main() {
	var projectID, clusterName, region, jobFilePath string
	flag.StringVar(&projectID, "project_id", "", "Cloud Project ID, used for creating resources.")
	flag.StringVar(&region, "region", "", "Region that resources should be created in.")
	flag.StringVar(&clusterName, "cluster_name", "", "Name of Cloud Dataproc cluster to create.")
	flag.StringVar(&jobFilePath, "job_file_path", "", "Path to job file in GCS.")
	flag.Parse()

	ctx := context.Background()

	// Create the cluster client.
	endpoint := fmt.Sprintf("%s-dataproc.googleapis.com:443", region)
	clusterClient, err := dataproc.NewClusterControllerClient(ctx, option.WithEndpoint(endpoint))
	if err != nil {
		log.Fatalf("error creating the cluster client: %s\n", err)
	}

	// Create the cluster config.
	createReq := &dataprocpb.CreateClusterRequest{
		ProjectId: projectID,
		Region:    region,
		Cluster: &dataprocpb.Cluster{
			ProjectId:   projectID,
			ClusterName: clusterName,
			Config: &dataprocpb.ClusterConfig{
				MasterConfig: &dataprocpb.InstanceGroupConfig{
					NumInstances:   1,
					MachineTypeUri: "n1-standard-2",
				},
				WorkerConfig: &dataprocpb.InstanceGroupConfig{
					NumInstances:   2,
					MachineTypeUri: "n1-standard-2",
				},
			},
		},
	}

	// Create the cluster.
	createOp, err := clusterClient.CreateCluster(ctx, createReq)
	if err != nil {
		log.Fatalf("error submitting the cluster creation request: %v\n", err)
	}

	createResp, err := createOp.Wait(ctx)
	if err != nil {
		log.Fatalf("error creating the cluster: %v\n", err)
	}

	// Defer cluster deletion.
	defer func() {
		dReq := &dataprocpb.DeleteClusterRequest{
			ProjectId:   projectID,
			Region:      region,
			ClusterName: clusterName,
		}
		deleteOp, err := clusterClient.DeleteCluster(ctx, dReq)
		deleteOp.Wait(ctx)
		if err != nil {
			fmt.Printf("error deleting cluster %q: %v\n", clusterName, err)
			return
		}
		fmt.Printf("Cluster %q successfully deleted\n", clusterName)
	}()

	// Output a success message.
	fmt.Printf("Cluster created successfully: %q\n", createResp.ClusterName)

	// Create the job client.
	jobClient, err := dataproc.NewJobControllerClient(ctx, option.WithEndpoint(endpoint))

	// Create the job config.
	submitJobReq := &dataprocpb.SubmitJobRequest{
		ProjectId: projectID,
		Region:    region,
		Job: &dataprocpb.Job{
			Placement: &dataprocpb.JobPlacement{
				ClusterName: clusterName,
			},
			TypeJob: &dataprocpb.Job_PysparkJob{
				PysparkJob: &dataprocpb.PySparkJob{
					MainPythonFileUri: jobFilePath,
				},
			},
		},
	}

	submitJobOp, err := jobClient.SubmitJobAsOperation(ctx, submitJobReq)
	if err != nil {
		fmt.Printf("error with request to submitting job: %v\n", err)
		return
	}

	submitJobResp, err := submitJobOp.Wait(ctx)
	if err != nil {
		fmt.Printf("error submitting job: %v\n", err)
		return
	}

	re := regexp.MustCompile("gs://(.+?)/(.+)")
	matches := re.FindStringSubmatch(submitJobResp.DriverOutputResourceUri)

	if len(matches) < 3 {
		fmt.Printf("regex error: %s\n", submitJobResp.DriverOutputResourceUri)
		return
	}

	// Dataproc job outget gets saved to a GCS bucket allocated to it.
	storageClient, err := storage.NewClient(ctx)
	if err != nil {
		fmt.Printf("error creating storage client: %v\n", err)
		return
	}

	obj := fmt.Sprintf("%s.000000000", matches[2])
	reader, err := storageClient.Bucket(matches[1]).Object(obj).NewReader(ctx)
	if err != nil {
		fmt.Printf("error reading job output: %v\n", err)
		return
	}

	defer reader.Close()

	body, err := io.ReadAll(reader)
	if err != nil {
		fmt.Printf("could not read output from Dataproc Job: %v\n", err)
		return
	}

	fmt.Printf("Job finished successfully: %s", body)
}

Java

クライアントライブラリをインストールします。詳細については、Java 開発環境の設定をご覧ください。
認証の設定

サンプル GitHub コードのクローンを作成して実行します。

/* This quickstart sample walks a user through creating a Cloud Dataproc
 * cluster, submitting a PySpark job from Google Cloud Storage to the
 * cluster, reading the output of the job and deleting the cluster, all
 * using the Java client library.
 *
 * Usage:
 *     mvn clean package -DskipTests
 *
 *     mvn exec:java -Dexec.args="<PROJECT_ID> <REGION> <CLUSTER_NAME> <GCS_JOB_FILE_PATH>"
 *
 *     You can also set these arguments in the main function instead of providing them via the CLI.
 */

import com.google.api.gax.longrunning.OperationFuture;
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.InstanceGroupConfig;
import com.google.cloud.dataproc.v1.Job;
import com.google.cloud.dataproc.v1.JobControllerClient;
import com.google.cloud.dataproc.v1.JobControllerSettings;
import com.google.cloud.dataproc.v1.JobMetadata;
import com.google.cloud.dataproc.v1.JobPlacement;
import com.google.cloud.dataproc.v1.PySparkJob;
import com.google.cloud.storage.Blob;
import com.google.cloud.storage.Storage;
import com.google.cloud.storage.StorageOptions;
import com.google.protobuf.Empty;
import java.io.IOException;
import java.util.concurrent.ExecutionException;
import java.util.regex.Matcher;
import java.util.regex.Pattern;

public class Quickstart {

  public static void quickstart(
      String projectId, String region, String clusterName, String jobFilePath)
      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 job controller client.
    JobControllerSettings jobControllerSettings =
        JobControllerSettings.newBuilder().setEndpoint(myEndpoint).build();

    // Create both a cluster controller client and job controller client with the
    // configured settings. The client only needs 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);
        JobControllerClient jobControllerClient =
            JobControllerClient.create(jobControllerSettings)) {
      // 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)
              .build();
      // Create the cluster object with the desired cluster config.
      Cluster cluster =
          Cluster.newBuilder().setClusterName(clusterName).setConfig(clusterConfig).build();

      // Create the Cloud Dataproc cluster.
      OperationFuture<Cluster, ClusterOperationMetadata> createClusterAsyncRequest =
          clusterControllerClient.createClusterAsync(projectId, region, cluster);
      Cluster clusterResponse = createClusterAsyncRequest.get();
      System.out.println(
          String.format("Cluster created successfully: %s", clusterResponse.getClusterName()));

      // Configure the settings for our job.
      JobPlacement jobPlacement = JobPlacement.newBuilder().setClusterName(clusterName).build();
      PySparkJob pySparkJob = PySparkJob.newBuilder().setMainPythonFileUri(jobFilePath).build();
      Job job = Job.newBuilder().setPlacement(jobPlacement).setPysparkJob(pySparkJob).build();

      // Submit an asynchronous request to execute the job.
      OperationFuture<Job, JobMetadata> submitJobAsOperationAsyncRequest =
          jobControllerClient.submitJobAsOperationAsync(projectId, region, job);
      Job jobResponse = submitJobAsOperationAsyncRequest.get();

      // Print output from Google Cloud Storage.
      Matcher matches =
          Pattern.compile("gs://(.*?)/(.*)").matcher(jobResponse.getDriverOutputResourceUri());
      matches.matches();

      Storage storage = StorageOptions.getDefaultInstance().getService();
      Blob blob = storage.get(matches.group(1), String.format("%s.000000000", matches.group(2)));

      System.out.println(
          String.format("Job finished successfully: %s", new String(blob.getContent())));

      // Delete the cluster.
      OperationFuture<Empty, ClusterOperationMetadata> deleteClusterAsyncRequest =
          clusterControllerClient.deleteClusterAsync(projectId, region, clusterName);
      deleteClusterAsyncRequest.get();
      System.out.println(String.format("Cluster \"%s\" successfully deleted.", clusterName));

    } catch (ExecutionException e) {
      System.err.println(String.format("quickstart: %s ", e.getMessage()));
    }
  }

  public static void main(String... args) throws IOException, InterruptedException {
    if (args.length != 4) {
      System.err.println(
          "Insufficient number of parameters provided. Please make sure a "
              + "PROJECT_ID, REGION, CLUSTER_NAME and JOB_FILE_PATH are provided, in this order.");
      return;
    }

    String projectId = args[0]; // project-id of project to create the cluster in
    String region = args[1]; // region to create the cluster
    String clusterName = args[2]; // name of the cluster
    String jobFilePath = args[3]; // location in GCS of the PySpark job

    quickstart(projectId, region, clusterName, jobFilePath);
  }
}

Node.js

クライアントライブラリをインストールします。詳細については、Node.js 開発環境の設定をご覧ください。
認証の設定

サンプル GitHub コードのクローンを作成して実行します。

// This quickstart sample walks a user through creating a Dataproc
// cluster, submitting a PySpark job from Google Cloud Storage to the
// cluster, reading the output of the job and deleting the cluster, all
// using the Node.js client library.

'use strict';

function main(projectId, region, clusterName, jobFilePath) {
  const dataproc = require('@google-cloud/dataproc');
  const {Storage} = require('@google-cloud/storage');

  // Create a cluster client with the endpoint set to the desired cluster region
  const clusterClient = new dataproc.v1.ClusterControllerClient({
    apiEndpoint: `${region}-dataproc.googleapis.com`,
    projectId: projectId,
  });

  // Create a job client with the endpoint set to the desired cluster region
  const jobClient = new dataproc.v1.JobControllerClient({
    apiEndpoint: `${region}-dataproc.googleapis.com`,
    projectId: projectId,
  });

  async function quickstart() {
    // Create the cluster config
    const cluster = {
      projectId: projectId,
      region: region,
      cluster: {
        clusterName: clusterName,
        config: {
          masterConfig: {
            numInstances: 1,
            machineTypeUri: 'n1-standard-2',
          },
          workerConfig: {
            numInstances: 2,
            machineTypeUri: 'n1-standard-2',
          },
        },
      },
    };

    // Create the cluster
    const [operation] = await clusterClient.createCluster(cluster);
    const [response] = await operation.promise();

    // Output a success message
    console.log(`Cluster created successfully: ${response.clusterName}`);

    const job = {
      projectId: projectId,
      region: region,
      job: {
        placement: {
          clusterName: clusterName,
        },
        pysparkJob: {
          mainPythonFileUri: jobFilePath,
        },
      },
    };

    const [jobOperation] = await jobClient.submitJobAsOperation(job);
    const [jobResponse] = await jobOperation.promise();

    const matches =
      jobResponse.driverOutputResourceUri.match('gs://(.*?)/(.*)');

    const storage = new Storage();

    const output = await storage
      .bucket(matches[1])
      .file(`${matches[2]}.000000000`)
      .download();

    // Output a success message.
    console.log(`Job finished successfully: ${output}`);

    // Delete the cluster once the job has terminated.
    const deleteClusterReq = {
      projectId: projectId,
      region: region,
      clusterName: clusterName,
    };

    const [deleteOperation] =
      await clusterClient.deleteCluster(deleteClusterReq);
    await deleteOperation.promise();

    // Output a success message
    console.log(`Cluster ${clusterName} successfully deleted.`);
  }

  quickstart();
}

const args = process.argv.slice(2);

if (args.length !== 4) {
  console.log(
    'Insufficient number of parameters provided. Please make sure a ' +
      'PROJECT_ID, REGION, CLUSTER_NAME and JOB_FILE_PATH are provided, in this order.'
  );
}

main(...args);

Python

クライアントライブラリをインストールします。詳細については、Python 開発環境の設定をご覧ください。
認証の設定

サンプル GitHub コードのクローンを作成して実行します。

"""
This quickstart sample walks a user through creating a Cloud Dataproc
cluster, submitting a PySpark job from Google Cloud Storage to the
cluster, reading the output of the job and deleting the cluster, all
using the Python client library.

Usage:
    python quickstart.py --project_id <PROJECT_ID> --region <REGION> \
        --cluster_name <CLUSTER_NAME> --job_file_path <GCS_JOB_FILE_PATH>
"""

import argparse
import re

from google.cloud import dataproc_v1 as dataproc
from google.cloud import storage


def quickstart(project_id, region, cluster_name, job_file_path):
    # Create the cluster client.
    cluster_client = dataproc.ClusterControllerClient(
        client_options={"api_endpoint": "{}-dataproc.googleapis.com:443".format(region)}
    )

    # Create the cluster config.
    cluster = {
        "project_id": project_id,
        "cluster_name": cluster_name,
        "config": {
            "master_config": {
                "num_instances": 1,
                "machine_type_uri": "n1-standard-2",
                "disk_config": {"boot_disk_size_gb": 100},
            },
            "worker_config": {
                "num_instances": 2,
                "machine_type_uri": "n1-standard-2",
                "disk_config": {"boot_disk_size_gb": 100},
            },
        },
    }

    # Create the cluster.
    operation = cluster_client.create_cluster(
        request={"project_id": project_id, "region": region, "cluster": cluster}
    )
    result = operation.result()

    print("Cluster created successfully: {}".format(result.cluster_name))

    # Create the job client.
    job_client = dataproc.JobControllerClient(
        client_options={"api_endpoint": "{}-dataproc.googleapis.com:443".format(region)}
    )

    # Create the job config.
    job = {
        "placement": {"cluster_name": cluster_name},
        "pyspark_job": {"main_python_file_uri": job_file_path},
    }

    operation = job_client.submit_job_as_operation(
        request={"project_id": project_id, "region": region, "job": job}
    )
    response = operation.result()

    # Dataproc job output gets saved to the Google Cloud Storage bucket
    # allocated to the job. Use a regex to obtain the bucket and blob info.
    matches = re.match("gs://(.*?)/(.*)", response.driver_output_resource_uri)

    output = (
        storage.Client()
        .get_bucket(matches.group(1))
        .blob(f"{matches.group(2)}.000000000")
        .download_as_bytes()
        .decode("utf-8")
    )

    print(f"Job finished successfully: {output}")

    # Delete the cluster once the job has terminated.
    operation = cluster_client.delete_cluster(
        request={
            "project_id": project_id,
            "region": region,
            "cluster_name": cluster_name,
        }
    )
    operation.result()

    print("Cluster {} successfully deleted.".format(cluster_name))


if __name__ == "__main__":
    parser = argparse.ArgumentParser(
        description=__doc__,
        formatter_class=argparse.RawDescriptionHelpFormatter,
    )
    parser.add_argument(
        "--project_id",
        type=str,
        required=True,
        help="Project to use for creating resources.",
    )
    parser.add_argument(
        "--region",
        type=str,
        required=True,
        help="Region where the resources should live.",
    )
    parser.add_argument(
        "--cluster_name",
        type=str,
        required=True,
        help="Name to use for creating a cluster.",
    )
    parser.add_argument(
        "--job_file_path",
        type=str,
        required=True,
        help="Job in GCS to execute against the cluster.",
    )

    args = parser.parse_args()
    quickstart(args.project_id, args.region, args.cluster_name, args.job_file_path)

出力を表示する

このコードでは、Cloud Storage のデフォルトの Managed Service for Apache Spark ステージングバケットにジョブドライバのログが出力されます。プロジェクトの Managed Service for Apache Spark の [Jobs] ページで、 Google Cloud コンソールからジョブドライバ出力を表示できます。 Google Cloud [ジョブ ID] をクリックすると、[ジョブの詳細] ページにジョブの出力が表示されます。

次のステップ

Managed Service for Apache Spark Cloud クライアントライブラリ追加リソースをご覧ください。

クライアント ライブラリを使用してクラスタを作成する

始める前に

コードの実行

Go

Java

Node.js

Python

出力を表示する

次のステップ

クライアントライブラリを使用してクラスタを作成する