Create and run a job that uses GPUs

This document explains how to create and run a job that uses a graphics processing unit (GPU).

When you create a Batch job, you can optionally add one or more GPUs to the VMs running it. Common use cases for jobs that use GPUs include intensive data processing and machine learning (ML) workloads.

Before you begin

  1. If you haven't used Batch before, review Get started with Batch and enable Batch by completing the prerequisites for projects and users.

  2. To get the permissions that you need to create a job, ask your administrator to grant you the following IAM roles:

    For more information about granting roles, see Manage access to projects, folders, and organizations.

    You might also be able to get the required permissions through custom roles or other predefined roles.

Create a job that uses GPUs

To create a job that uses GPUs, do the following:

  1. Review the Requirements for a job to use GPUs section to determine the methods you can use to create your job.
  2. Create a job with the methods you selected. For examples of how to create a job using the recommended methods, see the Create an example job that uses GPUs section.

Requirements for a job to use GPUs

To use GPUs, a job must do all of the following:

After you've determined how to meet these requirements for your job, you also need to define job's GPUs and location. A job's VMs can each use one or more GPUs of the type that you specify. The allowed locations for the job's VMs (or if undefined, the location of the job) must have the specified type of GPUs. For more information about defining the GPU type, GPU number, and a valid location for a job, see the examples.

Install GPU drivers

To install the required GPU drivers, select one of the following methods:

Define compatible VM resources

If your job defines any of the VM resources (any of the instances[] subfields) other than GPUs, you must define those VM resources in a compatible way.

To define the resources for a job's VMs, including any GPUs, you can only use of the following methods:

  • Define resources directly (recommended): As shown in the examples, to define the resources for a job's VMs directly, use the policy field.
  • Define resources in a template: Define the resource's for a job's VMs by specifying a Compute Engine instance template.

Additionally, all of the resources that you define must be compatible with the type and number of the GPUs for the job. For more information about the VM resources that you can use with GPUs, see GPU platforms in the Compute Engine documentation.

Create an example job that uses GPUs

You can create a job that uses GPUs using the gcloud CLI, Batch API, Java, or Python.

gcloud

  1. Create a JSON file that specifies the job's configuration details, the type and count subfields of the accelerators[] field, and a location that has those types of GPUs.

    For example, to create a basic script job that uses GPUs, automatically installs the required GPU drivers, and specifies the allowed locations for the job's VMs, create a JSON file with the following contents:

    {
    "taskGroups": [
        {
            "taskSpec": {
                "runnables": [
                    {
                        "script": {
                            "text": "echo Hello world from task ${BATCH_TASK_INDEX}."
                        }
                    }
                ]
            },
            "taskCount": 3,
            "parallelism": 1
        }
    ],
    "allocationPolicy": {
        "instances": [
            {
                "installGpuDrivers": INSTALL_GPU_DRIVERS,
                "policy": {
                    "accelerators": [
                        {
                            "type": "GPU_TYPE",
                            "count": GPU_COUNT
                        }
                    ]
                }
            }
        ],
        "location": {
            "allowedLocations": [
                "ALLOWED_LOCATIONS"
            ]
        }
    }
}

    Replace the following:

    • INSTALL_GPU_DRIVERS: Optional. When set to true, Batch fetches the drivers required for the GPU type that you specify in the policy field from a third-party location, and Batch installs them on your behalf. If you set this field to false (default), you need to install GPU drivers manually to use any GPUs for this job.

    • GPU_TYPE: the GPU type. You can view a list of the available GPU types by using the gcloud compute accelerator-types list command.

    • GPU_COUNT: the number of GPUs of the specified type.

    • ALLOWED_LOCATIONS: Optional. The locations where the VM instances for your job are allowed to run—for example, regions/us-central1, zones/us-central1-a allows the zone us-central1-a. If you specify an allowed location, you must select the region and, optionally, one or more zones. The locations that you choose must have the GPU types you want for this job. Otherwise, if you omit this field, the job's location must have the GPU types. For more information, see the allowedLocations[] field.

  2. To create and run the job, use the gcloud batch jobs submit command:

    gcloud batch jobs submit JOB_NAME \
    --location LOCATION \
    --config JSON_CONFIGURATION_FILE

    Replace the following:

    • JOB_NAME: the name of the job.

    • LOCATION: the location of the job.

    • JSON_CONFIGURATION_FILE: the path for a JSON file with the job's configuration details.

API

Make a POST request to the jobs.create method that specifies the job's configuration details, the type and count subfields of the accelerators[] field, and a location that has those types of GPUs.

For example, to create a basic script job that uses GPUs, automatically installs the required GPU drivers, and specifies the allowed locations for the job's VMs, create a JSON file with the following contents:

POST https://batch.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION/jobs?job_id=JOB_NAME

{
    "taskGroups": [
        {
            "taskSpec": {
                "runnables": [
                    {
                        "script": {
                            "text": "echo Hello world from task ${BATCH_TASK_INDEX}."
                        }
                    }
                ]
            },
            "taskCount": 3,
            "parallelism": 1
        }
    ],
    "allocationPolicy": {
        "instances": [
            {
                "installGpuDrivers": INSTALL_GPU_DRIVERS,
                "policy": {
                    "accelerators": [
                        {
                            "type": "GPU_TYPE",
                            "count": GPU_COUNT
                        }
                    ]
                }
            }
        ],
        "location": {
            "allowedLocations": [
                "ALLOWED_LOCATIONS"
            ]
        }
    }
}

Replace the following:

  • PROJECT_ID: the project ID of your project.

  • LOCATION: the location of the job.

  • JOB_NAME: the name of the job.

  • INSTALL_GPU_DRIVERS: Optional. When set to true, Batch fetches the drivers required for the GPU type that you specify in the policy field from a third-party location, and Batch installs them on your behalf. If you set this field to false (default), you need to install GPU drivers manually to use any GPUs for this job.

  • GPU_TYPE: the GPU type. You can view a list of the available GPU types by using the gcloud compute accelerator-types list command.

  • GPU_COUNT: the number of GPUs of the specified type.

  • ALLOWED_LOCATIONS: Optional. The locations where the VM instances for your job are allowed to run—for example, regions/us-central1, zones/us-central1-a allows the zone us-central1-a. If you specify an allowed location, you must select the region and, optionally, one or more zones. The locations that you choose must have the GPU types you want for this job. Otherwise, if you omit this field, the job's location must have the GPU types. For more information, see the allowedLocations[] field.

Java

To create a job with GPUs using Java, select one of the following options based on the machine type for your GPU model:

Create a job that uses GPUs with accelerator-optimized VMs

To use GPUs with accelerator-optimized VMs, just specify the machine type that you want for the job's VMs:


import com.google.cloud.batch.v1.AllocationPolicy;
import com.google.cloud.batch.v1.AllocationPolicy.Accelerator;
import com.google.cloud.batch.v1.AllocationPolicy.InstancePolicy;
import com.google.cloud.batch.v1.AllocationPolicy.InstancePolicyOrTemplate;
import com.google.cloud.batch.v1.BatchServiceClient;
import com.google.cloud.batch.v1.CreateJobRequest;
import com.google.cloud.batch.v1.Job;
import com.google.cloud.batch.v1.LogsPolicy;
import com.google.cloud.batch.v1.Runnable;
import com.google.cloud.batch.v1.Runnable.Script;
import com.google.cloud.batch.v1.TaskGroup;
import com.google.cloud.batch.v1.TaskSpec;
import com.google.protobuf.Duration;
import java.io.IOException;
import java.util.concurrent.ExecutionException;
import java.util.concurrent.TimeUnit;
import java.util.concurrent.TimeoutException;

public class CreateGpuJob {

  public static void main(String[] args)
      throws IOException, ExecutionException, InterruptedException, TimeoutException {
    // TODO(developer): Replace these variables before running the sample.
    // Project ID or project number of the Google Cloud project you want to use.
    String projectId = "YOUR_PROJECT_ID";
    // Name of the region you want to use to run the job. Regions that are
    // available for Batch are listed on: https://cloud.google.com/batch/docs/get-started#locations
    String region = "europe-central2";
    // The name of the job that will be created.
    // It needs to be unique for each project and region pair.
    String jobName = "JOB_NAME";
    // Optional. When set to true, Batch fetches the drivers required for the GPU type
    // that you specify in the policy field from a third-party location,
    // and Batch installs them on your behalf. If you set this field to false (default),
    // you need to install GPU drivers manually to use any GPUs for this job.
    boolean installGpuDrivers = false;
    // Accelerator-optimized machine types are available to Batch jobs. See the list
    // of available types on: https://cloud.google.com/compute/docs/accelerator-optimized-machines
    String machineType = "g2-standard-4";

    createGpuJob(projectId, region, jobName, installGpuDrivers, machineType);
  }

  // Create a job that uses GPUs
  public static Job createGpuJob(String projectId, String region, String jobName,
                                  boolean installGpuDrivers, String machineType)
      throws IOException, ExecutionException, InterruptedException, TimeoutException {
    // Initialize client that will be used to send requests. This client only needs to be created
    // once, and can be reused for multiple requests.
    try (BatchServiceClient batchServiceClient = BatchServiceClient.create()) {
      // Define what will be done as part of the job.
      Runnable runnable =
          Runnable.newBuilder()
              .setScript(
                  Script.newBuilder()
                      .setText(
                          "echo Hello world! This is task ${BATCH_TASK_INDEX}. "
                                  + "This job has a total of ${BATCH_TASK_COUNT} tasks.")
                      // You can also run a script from a file. Just remember, that needs to be a
                      // script that's already on the VM that will be running the job.
                      // Using setText() and setPath() is mutually exclusive.
                      // .setPath("/tmp/test.sh")
                      .build())
              .build();

      TaskSpec task = TaskSpec.newBuilder()
                  // Jobs can be divided into tasks. In this case, we have only one task.
                  .addRunnables(runnable)
                  .setMaxRetryCount(2)
                  .setMaxRunDuration(Duration.newBuilder().setSeconds(3600).build())
                  .build();

      // Tasks are grouped inside a job using TaskGroups.
      // Currently, it's possible to have only one task group.
      TaskGroup taskGroup = TaskGroup.newBuilder()
          .setTaskCount(3)
          .setParallelism(1)
          .setTaskSpec(task)
          .build();

      // Policies are used to define on what kind of virtual machines the tasks will run.
      // Read more about machine types here: https://cloud.google.com/compute/docs/machine-types
      InstancePolicy instancePolicy =
          InstancePolicy.newBuilder().setMachineType(machineType).build();  

      // Policies are used to define on what kind of virtual machines the tasks will run on.
      AllocationPolicy allocationPolicy =
          AllocationPolicy.newBuilder()
              .addInstances(
                  InstancePolicyOrTemplate.newBuilder()
                      .setInstallGpuDrivers(installGpuDrivers)
                      .setPolicy(instancePolicy)
                      .build())
              .build();

      Job job =
          Job.newBuilder()
              .addTaskGroups(taskGroup)
              .setAllocationPolicy(allocationPolicy)
              .putLabels("env", "testing")
              .putLabels("type", "script")
              // We use Cloud Logging as it's an out of the box available option.
              .setLogsPolicy(
                  LogsPolicy.newBuilder().setDestination(LogsPolicy.Destination.CLOUD_LOGGING))
              .build();

      CreateJobRequest createJobRequest =
          CreateJobRequest.newBuilder()
              // The job's parent is the region in which the job will run.
              .setParent(String.format("projects/%s/locations/%s", projectId, region))
              .setJob(job)
              .setJobId(jobName)
              .build();

      Job result =
          batchServiceClient
              .createJobCallable()
              .futureCall(createJobRequest)
              .get(5, TimeUnit.MINUTES);

      System.out.printf("Successfully created the job: %s", result.getName());

      return result;
    }
  }
}

Create a job that uses GPUs with N1 VMs

To use GPUs with N1 VMs, you need to specify the number and type of GPUs that you want for each of the job's VMs:


import com.google.cloud.batch.v1.AllocationPolicy;
import com.google.cloud.batch.v1.AllocationPolicy.Accelerator;
import com.google.cloud.batch.v1.AllocationPolicy.InstancePolicy;
import com.google.cloud.batch.v1.AllocationPolicy.InstancePolicyOrTemplate;
import com.google.cloud.batch.v1.BatchServiceClient;
import com.google.cloud.batch.v1.CreateJobRequest;
import com.google.cloud.batch.v1.Job;
import com.google.cloud.batch.v1.LogsPolicy;
import com.google.cloud.batch.v1.Runnable;
import com.google.cloud.batch.v1.Runnable.Script;
import com.google.cloud.batch.v1.TaskGroup;
import com.google.cloud.batch.v1.TaskSpec;
import com.google.protobuf.Duration;
import java.io.IOException;
import java.util.concurrent.ExecutionException;
import java.util.concurrent.TimeUnit;
import java.util.concurrent.TimeoutException;

public class CreateGpuJobN1 {

  public static void main(String[] args)
      throws IOException, ExecutionException, InterruptedException, TimeoutException {
    // TODO(developer): Replace these variables before running the sample.
    // Project ID or project number of the Google Cloud project you want to use.
    String projectId = "YOUR_PROJECT_ID";
    // Name of the region you want to use to run the job. Regions that are
    // available for Batch are listed on: https://cloud.google.com/batch/docs/get-started#locations
    String region = "europe-central2";
    // The name of the job that will be created.
    // It needs to be unique for each project and region pair.
    String jobName = "JOB_NAME";
    // Optional. When set to true, Batch fetches the drivers required for the GPU type
    // that you specify in the policy field from a third-party location,
    // and Batch installs them on your behalf. If you set this field to false (default),
    // you need to install GPU drivers manually to use any GPUs for this job.
    boolean installGpuDrivers = false;
    // The GPU type. You can view a list of the available GPU types
    // by using the `gcloud compute accelerator-types list` command.
    String gpuType = "nvidia-tesla-t4";
    // The number of GPUs of the specified type.
    int gpuCount = 2;

    createGpuJob(projectId, region, jobName, installGpuDrivers, gpuType, gpuCount);
  }

  // Create a job that uses GPUs
  public static Job createGpuJob(String projectId, String region, String jobName,
                                  boolean installGpuDrivers, String gpuType, int gpuCount)
      throws IOException, ExecutionException, InterruptedException, TimeoutException {
    // Initialize client that will be used to send requests. This client only needs to be created
    // once, and can be reused for multiple requests.
    try (BatchServiceClient batchServiceClient = BatchServiceClient.create()) {
      // Define what will be done as part of the job.
      Runnable runnable =
          Runnable.newBuilder()
              .setScript(
                  Script.newBuilder()
                      .setText(
                          "echo Hello world! This is task ${BATCH_TASK_INDEX}. "
                                  + "This job has a total of ${BATCH_TASK_COUNT} tasks.")
                      // You can also run a script from a file. Just remember, that needs to be a
                      // script that's already on the VM that will be running the job.
                      // Using setText() and setPath() is mutually exclusive.
                      // .setPath("/tmp/test.sh")
                      .build())
              .build();

      TaskSpec task = TaskSpec.newBuilder()
                  // Jobs can be divided into tasks. In this case, we have only one task.
                  .addRunnables(runnable)
                  .setMaxRetryCount(2)
                  .setMaxRunDuration(Duration.newBuilder().setSeconds(3600).build())
                  .build();

      // Tasks are grouped inside a job using TaskGroups.
      // Currently, it's possible to have only one task group.
      TaskGroup taskGroup = TaskGroup.newBuilder()
          .setTaskCount(3)
          .setParallelism(1)
          .setTaskSpec(task)
          .build();

      // Accelerator describes Compute Engine accelerators to be attached to the VM.
      Accelerator accelerator = Accelerator.newBuilder()
          .setType(gpuType)
          .setCount(gpuCount)
          .build();

      // Policies are used to define on what kind of virtual machines the tasks will run on.
      AllocationPolicy allocationPolicy =
          AllocationPolicy.newBuilder()
              .addInstances(
                  InstancePolicyOrTemplate.newBuilder()
                      .setInstallGpuDrivers(installGpuDrivers)
                      .setPolicy(InstancePolicy.newBuilder().addAccelerators(accelerator))
                      .build())
              .build();

      Job job =
          Job.newBuilder()
              .addTaskGroups(taskGroup)
              .setAllocationPolicy(allocationPolicy)
              .putLabels("env", "testing")
              .putLabels("type", "script")
              // We use Cloud Logging as it's an out of the box available option.
              .setLogsPolicy(
                  LogsPolicy.newBuilder().setDestination(LogsPolicy.Destination.CLOUD_LOGGING))
              .build();

      CreateJobRequest createJobRequest =
          CreateJobRequest.newBuilder()
              // The job's parent is the region in which the job will run.
              .setParent(String.format("projects/%s/locations/%s", projectId, region))
              .setJob(job)
              .setJobId(jobName)
              .build();

      Job result =
          batchServiceClient
              .createJobCallable()
              .futureCall(createJobRequest)
              .get(5, TimeUnit.MINUTES);

      System.out.printf("Successfully created the job: %s", result.getName());

      return result;
    }
  }
}

Python

To create a job with GPUs using Python, select one of the following options based on the machine type for your GPU model:

Create a job that uses GPUs with accelerator-optimized VMs

To use GPUs with accelerator-optimized VMs, just specify the machine type that you want for the job's VMs:

from google.cloud import batch_v1


def create_gpu_job(project_id: str, region: str, job_name: str) -> batch_v1.Job:
    """
    This method shows how to create a sample Batch Job that will run
    a simple command on Cloud Compute instances on GPU machines.

    Args:
        project_id: project ID or project number of the Cloud project you want to use.
        region: name of the region you want to use to run the job. Regions that are
            available for Batch are listed on: https://cloud.google.com/batch/docs/get-started#locations
        job_name: the name of the job that will be created.
            It needs to be unique for each project and region pair.

    Returns:
        A job object representing the job created.
    """
    client = batch_v1.BatchServiceClient()

    # Define what will be done as part of the job.
    task = batch_v1.TaskSpec()
    runnable = batch_v1.Runnable()
    runnable.script = batch_v1.Runnable.Script()
    runnable.script.text = "echo Hello world! This is task ${BATCH_TASK_INDEX}. This job has a total of ${BATCH_TASK_COUNT} tasks."
    # You can also run a script from a file. Just remember, that needs to be a script that's
    # already on the VM that will be running the job. Using runnable.script.text and runnable.script.path is mutually
    # exclusive.
    # runnable.script.path = '/tmp/test.sh'
    task.runnables = [runnable]

    # We can specify what resources are requested by each task.
    resources = batch_v1.ComputeResource()
    resources.cpu_milli = 2000  # in milliseconds per cpu-second. This means the task requires 2 whole CPUs.
    resources.memory_mib = 16  # in MiB
    task.compute_resource = resources

    task.max_retry_count = 2
    task.max_run_duration = "3600s"

    # Tasks are grouped inside a job using TaskGroups.
    # Currently, it's possible to have only one task group.
    group = batch_v1.TaskGroup()
    group.task_count = 4
    group.task_spec = task

    # Policies are used to define on what kind of virtual machines the tasks will run on.
    # In this case, we tell the system to use "g2-standard-4" machine type.
    # Read more about machine types here: https://cloud.google.com/compute/docs/machine-types
    policy = batch_v1.AllocationPolicy.InstancePolicy()
    policy.machine_type = "g2-standard-4"

    instances = batch_v1.AllocationPolicy.InstancePolicyOrTemplate()
    instances.policy = policy
    instances.install_gpu_drivers = True
    allocation_policy = batch_v1.AllocationPolicy()
    allocation_policy.instances = [instances]

    job = batch_v1.Job()
    job.task_groups = [group]
    job.allocation_policy = allocation_policy
    job.labels = {"env": "testing", "type": "container"}
    # We use Cloud Logging as it's an out of the box available option
    job.logs_policy = batch_v1.LogsPolicy()
    job.logs_policy.destination = batch_v1.LogsPolicy.Destination.CLOUD_LOGGING

    create_request = batch_v1.CreateJobRequest()
    create_request.job = job
    create_request.job_id = job_name
    # The job's parent is the region in which the job will run
    create_request.parent = f"projects/{project_id}/locations/{region}"

    return client.create_job(create_request)

Create a job that uses GPUs with N1 VMs

To use GPUs with N1 VMs, you need to specify the number and type of GPUs that you want for each of the job's VMs:

from google.cloud import batch_v1


def create_gpu_job(
    project_id: str, region: str, zone: str, job_name: str
) -> batch_v1.Job:
    """
    This method shows how to create a sample Batch Job that will run
    a simple command on Cloud Compute instances on GPU machines.

    Args:
        project_id: project ID or project number of the Cloud project you want to use.
        region: name of the region you want to use to run the job. Regions that are
            available for Batch are listed on: https://cloud.google.com/batch/docs/get-started#locations
        zone: name of the zone you want to use to run the job. Important in regard to GPUs availability.
            GPUs availability can be found here: https://cloud.google.com/compute/docs/gpus/gpu-regions-zones
        job_name: the name of the job that will be created.
            It needs to be unique for each project and region pair.

    Returns:
        A job object representing the job created.
    """
    client = batch_v1.BatchServiceClient()

    # Define what will be done as part of the job.
    task = batch_v1.TaskSpec()
    runnable = batch_v1.Runnable()
    runnable.script = batch_v1.Runnable.Script()
    runnable.script.text = "echo Hello world! This is task ${BATCH_TASK_INDEX}. This job has a total of ${BATCH_TASK_COUNT} tasks."
    # You can also run a script from a file. Just remember, that needs to be a script that's
    # already on the VM that will be running the job. Using runnable.script.text and runnable.script.path is mutually
    # exclusive.
    # runnable.script.path = '/tmp/test.sh'
    task.runnables = [runnable]

    # We can specify what resources are requested by each task.
    resources = batch_v1.ComputeResource()
    resources.cpu_milli = 2000  # in milliseconds per cpu-second. This means the task requires 2 whole CPUs.
    resources.memory_mib = 16  # in MiB
    task.compute_resource = resources

    task.max_retry_count = 2
    task.max_run_duration = "3600s"

    # Tasks are grouped inside a job using TaskGroups.
    # Currently, it's possible to have only one task group.
    group = batch_v1.TaskGroup()
    group.task_count = 4
    group.task_spec = task

    # Policies are used to define on what kind of virtual machines the tasks will run on.
    # Read more about machine types here: https://cloud.google.com/compute/docs/machine-types
    policy = batch_v1.AllocationPolicy.InstancePolicy()
    policy.machine_type = "n1-standard-16"

    accelerator = batch_v1.AllocationPolicy.Accelerator()
    # Note: not every accelerator is compatible with instance type
    # Read more here: https://cloud.google.com/compute/docs/gpus#t4-gpus
    accelerator.type_ = "nvidia-tesla-t4"
    accelerator.count = 1

    policy.accelerators = [accelerator]
    instances = batch_v1.AllocationPolicy.InstancePolicyOrTemplate()
    instances.policy = policy
    instances.install_gpu_drivers = True
    allocation_policy = batch_v1.AllocationPolicy()
    allocation_policy.instances = [instances]

    location = batch_v1.AllocationPolicy.LocationPolicy()
    location.allowed_locations = ["zones/us-central1-b"]
    allocation_policy.location = location

    job = batch_v1.Job()
    job.task_groups = [group]
    job.allocation_policy = allocation_policy
    job.labels = {"env": "testing", "type": "container"}
    # We use Cloud Logging as it's an out of the box available option
    job.logs_policy = batch_v1.LogsPolicy()
    job.logs_policy.destination = batch_v1.LogsPolicy.Destination.CLOUD_LOGGING

    create_request = batch_v1.CreateJobRequest()
    create_request.job = job
    create_request.job_id = job_name
    # The job's parent is the region in which the job will run
    create_request.parent = f"projects/{project_id}/locations/{region}"

    return client.create_job(create_request)

What's next