Create AI-optimized G4 VMs in bulk

General GPUs

This document explains how to create virtual machine (VM) instances in bulk that use the G4 machine series. To learn more about these accelerator-optimized machine types, see G4 series.

For more information about creating VMs in bulk, see About bulk creation of VMs in the Compute Engine documentation.

To learn about other ways to create VMs or clusters, see Deployment options overview.

Limitations

When you create G4 VMs in bulk, the following limitations apply:

• You don't receive sustained use discounts and flexible committed use discounts for instances that use a G4 machine type.
• You can only use a G4 machine type in certain regions and zones.
• You can't use Persistent Disk (regional or zonal) on an instance that uses a G4 machine type.
• The G4 machine type is only available on the AMD EPYC Turin 5th Generation platform.
• You can only use a Confidential VM with the g4-standard-48 machine type in limited regions and zones. All the limitations for Confidential VM running on the G4 machine type apply.
• You can't create G4 instances on sole-tenant nodes.
• You can't use Windows operating systems on g4-standard-384 instances.
• You can't attach Hyperdisk ML disks that were created before February 4, 2026 to G4 machine types.
• When creating G4 instances that have less than one GPU attached (fractional GPUs), don't use the --no-service-account or --no-scopes flags. To authenticate NVIDIA vGPU drivers, Compute Engine must verify the VM's identity. This process requires that service accounts are enabled.

If you want to use a compact placement policy, then review the restrictions for placement policies.

Before you begin

Before creating VMs in bulk, if you haven't already done so, complete the following steps:

1. Choose a consumption option: your choice of consumption option determines how you get and use GPU resources. To learn more, see Choose a consumption option.
2. Obtain capacity: the process to obtain capacity differs for each consumption option. To learn about the process to obtain capacity for your chosen consumption option, see Capacity overview.

Select the tab for how you plan to use the samples on this page:

Console

When you use the Google Cloud console to access Google Cloud services and APIs, you don't need to set up authentication.

gcloud

In the Google Cloud console, activate Cloud Shell.

Activate Cloud Shell

At the bottom of the Google Cloud console, a Cloud Shell session starts and displays a command-line prompt. Cloud Shell is a shell environment with the Google Cloud CLI already installed and with values already set for your current project. It can take a few seconds for the session to initialize.

REST

To use the REST API samples on this page in a local development environment, you use the credentials you provide to the gcloud CLI.

Install the Google Cloud CLI.

If you're using an external identity provider (IdP), you must first sign in to the gcloud CLI with your federated identity.

For more information, see Authenticate for using REST in the Google Cloud authentication documentation.

Required roles

To get the permissions that you need to create VMs in bulk, ask your administrator to grant you the Compute Instance Admin (v1) (roles/compute.instanceAdmin.v1) IAM role on the project. For more information about granting roles, see Manage access to projects, folders, and organizations.

This predefined role contains the permissions required to create VMs in bulk. To see the exact permissions that are required, expand the Required permissions section:

Required permissions

The following permissions are required to create VMs in bulk:

• compute.instances.create on the project
• To use a custom image to create the VM: compute.images.useReadOnly on the image
• To use a snapshot to create the VM: compute.snapshots.useReadOnly on the snapshot
• To use an instance template to create the VM: compute.instanceTemplates.useReadOnly on the instance template
• To specify a subnet for your VM: compute.subnetworks.use on the project or on the chosen subnet
• To specify a static IP address for the VM: compute.addresses.use on the project
• To assign an external IP address to the VM when using a VPC network: compute.subnetworks.useExternalIp on the project or on the chosen subnet
• To assign a legacy network to the VM: compute.networks.use on the project
• To assign an external IP address to the VM when using a legacy network: compute.networks.useExternalIp on the project
• To set VM instance metadata for the VM: compute.instances.setMetadata on the project
• To set tags for the VM: compute.instances.setTags on the VM
• To set labels for the VM: compute.instances.setLabels on the VM
• To set a service account for the VM to use: compute.instances.setServiceAccount on the VM
• To create a new disk for the VM: compute.disks.create on the project
• To attach an existing disk in read-only or read-write mode: compute.disks.use on the disk
• To attach an existing disk in read-only mode: compute.disks.useReadOnly on the disk

You might also be able to get these permissions with custom roles or other predefined roles.

Overview

Creating instances in bulk with the G4 machine type includes the following steps:

1. Optional: Create a compact placement policy
2. Create instances in bulk
3. Install GPU drivers
4. Optional: Enable NVIDIA Multi-Instance GPU mode

Optional: Create a compact placement policy

Tip: Decide whether to create a compact placement policy as follows:

• If you want to place your VMs in single or adjacent blocks, then specify VM placement by creating a compact placement policy and applying it when you create VMs.

You can specify VM placement by creating a compact placement policy. When you apply a compact placement policy to your VMs, Compute Engine makes best-effort attempts to create VMs that are as close to each other as possible. If your application is latency-sensitive and requires maximum compactness, then specify the maxDistance field (Preview) when you create a compact placement policy. A lower maxDistance value ensures closer VM placement, but it also increases the chance that some VMs won't be created.

To create a compact placement policy, select one of the following options:

gcloud

To create a compact placement policy, use the gcloud beta compute resource-policies create group-placement command:

gcloud beta compute resource-policies create group-placement POLICY_NAME \
    --collocation=collocated \
    --max-distance=MAX_DISTANCE \
    --region=REGION

Replace the following:

• POLICY_NAME: the name of the compact placement policy.
• MAX_DISTANCE: the maximum distance configuration for your VMs, which can either be specified (Preview, recommended) or unspecified (remove the --max-distance flag). Ensure that the maximum distance configuration supports the machine series and number of VMs that you plan to create. For more information, see About compact placement policies in the Compute Engine documentation. For example, a maximum distance of 3 places VMs in adjacent blocks and 2 places VMs in the same block.
• REGION: the region where you want to create the compact placement policy. Specify the region that contains the zone where you plan to create instances, and verify that the machine type that you want to use is available in that region. For information about regions, see GPU availability by regions and zones.

REST

To create a compact placement policy, make a POST request to the beta resourcePolicies.insert method. In the request body, include the collocation field set to COLLOCATED, and the maxDistance field.

POST https://compute.googleapis.com/compute/beta/projects/PROJECT_ID/regions/REGION/resourcePolicies
  {
    "name": "POLICY_NAME",
    "groupPlacementPolicy": {
      "collocation": "COLLOCATED",
      "maxDistance": MAX_DISTANCE
    }
  }

Replace the following:

• PROJECT_ID: your project ID.
• POLICY_NAME: the name of the compact placement policy.
• MAX_DISTANCE: the maximum distance configuration for your VMs, which can either be specified (Preview, recommended) or unspecified (remove the maxDistance field). Ensure that the maximum distance configuration supports the machine series and number of VMs that you plan to create. For more information, see About compact placement policies in the Compute Engine documentation. For example, a maximum distance of 3 places VMs in adjacent blocks and 2 places VMs in the same block.
• REGION: the region where you want to create the compact placement policy. Specify the region that contains the zone where you plan to create instances, and verify that the machine type that you want to use is available in that region. For information about regions, see GPU availability by regions and zones.

Create G4 VMs in bulk

To create G4 VMs in bulk, use one of the following methods.

The following commands also set the access scope for your instances. To simplify permissions management, Google recommends that you set the access scope on an instance to cloud-platform access and then use IAM roles to define what services the instance can access. For more information, see Scopes best practice.

gcloud

To create instances in bulk, use the gcloud compute instances bulk create command.

The parameters that you need to specify depend on the consumption option that you are using for this deployment. Select the tab that corresponds to your consumption option.

Reservation

G4 instances support the following types of reservations:

• On-demand reservations
• Standard future reservations

To consume these types of reservations, an instance must use the standard provisioning model. For example, use the following creation parameters.

Before you run the command, review all of the following steps to determine whether to include additional flags.

gcloud compute instances bulk create \
    --name-pattern=NAME_PATTERN \
    --count=COUNT \
     --machine-type=MACHINE_TYPE \
    --image-family=IMAGE_FAMILY \
    --image-project=IMAGE_PROJECT \
    --region=REGION \
    --boot-disk-type=hyperdisk-balanced \
    --boot-disk-size=DISK_SIZE \
    --scopes=cloud-platform \
    --provisioning-model=STANDARD \
    --maintenance-policy=TERMINATE \
    --reservation-affinity=RESERVATION_AFFINITY \
    --reservation=RESERVATION

Complete the following steps:

1. Replace the following:

   • NAME_PATTERN: the name pattern to use for the G4 instances. For example, using instance-# for the name pattern generates G4 instances with names such as instance-1 and instance-2, up to the number of G4 instances specified by --count.
   • COUNT: the number of G4 instances to create.
   • MACHINE_TYPE: the machine type to use for the G4 instance. For more information, see GPU machine types in the Compute Engine documentation.
   • IMAGE_FAMILY: the image family of the OS image that you want to use. For a list of all supported operating systems, see Supported operating systems.
   • IMAGE_PROJECT: the project ID of the OS image.
   • REGION: specify a region in which the machine type that you want to use is available. If you want to specify a compact placement policy, then you must use the same region as the compact placement policy. For information about regions, see GPU availability by regions and zones.
   • DISK_SIZE: the size of the boot disk in GB. For more information, see either Google Cloud Hyperdisk size limits or Persistent Disk size limits based on the type of disk.

   • RESERVATION_AFFINITY: which reservations the instance is allowed to consume, if possible. Specify one of the following:

     • any (Reservation capacity is optional): this is the default option. The instance tries to use capacity from an automatically consumed reservation, which only happens if matching reservation capacity is available. Otherwise, the instance uses on-demand capacity.

     Caution: When matching reservation capacity is unavailable, the instance uses the on-demand consumption option instead, which has different billing and capacity assurance. For more information, see Choose a consumption option.

     • specific (Reservation capacity is required): Require the instance to use capacity from a specifically targeted reservation. The instance isn't allowed to use on-demand capacity. If reservation capacity is unavailable, then the instance creation fails.

   • RESERVATION: the reservation to use. Specify one of the following based on the value you specified for RESERVATION_AFFINITY:

     • If you specified any, then replace RESERVATION with an empty string (""). Alternatively, you can remove the --reservation flag entirely.

     • If you specified specific, then replace RESERVATION with the following:

       projects/RESERVATION_OWNER_PROJECT_ID/reservations/RESERVATION_NAME

       Replace the following:

       • RESERVATION_OWNER_PROJECT_ID: the project ID of the project where the reservation was created.
       Tip: If the reservation exists in the current project, then you can omit projects/RESERVATION_OWNER_PROJECT_ID/reservations/ from the RESERVATION value.
       • RESERVATION_NAME: the name of the reservation.

2. By default, the instance automatically restarts if it crashes or Compute Engine stops it for a programmed stop, such as a maintenance event. If you want to disable automatic restart, then add the following flag:

   --no-restart-on-failure

3. Optional: If you chose to use a compact placement policy, then add the following flag to the command:

   --resource-policies=POLICY_NAME
   

   Replace the following:

   • POLICY_NAME: the name of the compact placement policy.

4. Optional: To attach Local SSD disks, then add the following --local-ssd flag for each Local SSD disk.

   Ensure that the number of Local SSD disks that you attach is allowed for your machine type; for details, see About Local SSD disks in the Compute Engine documentation. Specifically, G4 machine types use Titanium SSD disks.

   --local-ssd interface=INTERFACE_TYPE
   

   Replace the following:

   • INTERFACE_TYPE: the disk interface type that you want to use for the Local SSD disk. If your boot disk image has optimized NVMe drivers, then specify NVME. Specify SCSI for other images.

   After creating a VM with Local SSD disks, you must format and mount each device before you can use it.

5. Optional: To use NVIDIA RTX Virtual Workstation (vWS), add the following flag.

   When you create an instance that uses NVIDIA RTX Virtual Workstation, Compute Engine automatically adds a vWS license. For information about pricing for virtual workstations, see the GPU pricing page.

   --accelerator=type=nvidia-rtx-pro-6000-vws,count=VWS_ACCELERATOR_COUNT
   

   Replace VWS_ACCELERATOR_COUNT with the number of virtual GPUs that you want.

6. Run the command.

On-demand

Before you run the command, review all of the following steps to determine whether to include additional flags.

gcloud compute instances bulk create \
    --name-pattern=NAME_PATTERN \
    --count=COUNT \
     --machine-type=MACHINE_TYPE \
    --image-family=IMAGE_FAMILY \
    --image-project=IMAGE_PROJECT \
    --region=REGION \
    --boot-disk-type=hyperdisk-balanced \
    --boot-disk-size=DISK_SIZE \
    --scopes=cloud-platform \
    --provisioning-model=STANDARD \
    --maintenance-policy=TERMINATE \
    --reservation-affinity=RESERVATION_AFFINITY

Complete the following steps:

1. Replace the following:

   • NAME_PATTERN: the name pattern to use for the G4 instances. For example, using instance-# for the name pattern generates G4 instances with names such as instance-1 and instance-2, up to the number of G4 instances specified by --count.
   • COUNT: the number of G4 instances to create.
   • MACHINE_TYPE: the machine type to use for the G4 instance. For more information, see GPU machine types in the Compute Engine documentation.
   • IMAGE_FAMILY: the image family of the OS image that you want to use. For a list of all supported operating systems, see Supported operating systems.
   • IMAGE_PROJECT: the project ID of the OS image.
   • REGION: specify a region in which the machine type that you want to use is available. If you want to specify a compact placement policy, then you must use the same region as the compact placement policy. For information about regions, see GPU availability by regions and zones.
   • DISK_SIZE: the size of the boot disk in GB. For more information, see either Google Cloud Hyperdisk size limits or Persistent Disk size limits based on the type of disk.

   • RESERVATION_AFFINITY: which reservations the instance is allowed to consume, if possible. Specify one of the following:

     • any (Reservation capacity is optional): this is the default option. The instance tries to use capacity from an automatically consumed reservation, which only happens if matching reservation capacity is available. Otherwise, the instance uses on-demand capacity.

     Note: If you want using reservation capacity to be required, then follow the Reservation steps instead.

     • none (On-demand capacity is required): Block the instance from using any reservation capacity.

2. By default, the instance automatically restarts if it crashes or Compute Engine stops it for a programmed stop, such as a maintenance event. If you want to disable automatic restart, then add the following flag:

   --no-restart-on-failure

3. Optional: If you chose to use a compact placement policy, then add the following flag to the command:

   --resource-policies=POLICY_NAME
   

   Replace the following:

   • POLICY_NAME: the name of the compact placement policy.

4. Optional: To attach Local SSD disks, then add the following --local-ssd flag for each Local SSD disk.

   Ensure that the number of Local SSD disks that you attach is allowed for your machine type; for details, see About Local SSD disks in the Compute Engine documentation. Specifically, G4 machine types use Titanium SSD disks.

   --local-ssd interface=INTERFACE_TYPE
   

   Replace the following:

   • INTERFACE_TYPE: the disk interface type that you want to use for the Local SSD disk. If your boot disk image has optimized NVMe drivers, then specify NVME. Specify SCSI for other images.

   After creating a VM with Local SSD disks, you must format and mount each device before you can use it.

5. Optional: To use NVIDIA RTX Virtual Workstation (vWS), add the following flag.

   When you create an instance that uses NVIDIA RTX Virtual Workstation, Compute Engine automatically adds a vWS license. For information about pricing for virtual workstations, see the GPU pricing page.

   --accelerator=type=nvidia-rtx-pro-6000-vws,count=VWS_ACCELERATOR_COUNT
   

   Replace VWS_ACCELERATOR_COUNT with the number of virtual GPUs that you want.

6. Run the command.

Spot

Before you run the command, review all of the following steps to determine whether to include additional flags.

gcloud compute instances bulk create \
    --name-pattern=NAME_PATTERN \
    --count=COUNT \
     --machine-type=MACHINE_TYPE \
    --image-family=IMAGE_FAMILY \
    --image-project=IMAGE_PROJECT \
    --region=REGION \
    --boot-disk-type=hyperdisk-balanced \
    --boot-disk-size=DISK_SIZE \
    --scopes=cloud-platform \
    --provisioning-model=SPOT \
    --instance-termination-action=TERMINATION_ACTION \
    --maintenance-policy=TERMINATE \
    --no-restart-on-failure

Complete the following steps:

1. Replace the following:

   • NAME_PATTERN: the name pattern to use for the G4 instances. For example, using instance-# for the name pattern generates G4 instances with names such as instance-1 and instance-2, up to the number of G4 instances specified by --count.
   • COUNT: the number of G4 instances to create.
   • MACHINE_TYPE: the machine type to use for the G4 instance. For more information, see GPU machine types in the Compute Engine documentation.
   • IMAGE_FAMILY: the image family of the OS image that you want to use. For a list of all supported operating systems, see Supported operating systems.
   • IMAGE_PROJECT: the project ID of the OS image.
   • REGION: specify a region in which the machine type that you want to use is available. If you want to specify a compact placement policy, then you must use the same region as the compact placement policy. For information about regions, see GPU availability by regions and zones.
   • DISK_SIZE: the size of the boot disk in GB. For more information, see either Google Cloud Hyperdisk size limits or Persistent Disk size limits based on the type of disk.

   • TERMINATION_ACTION: the action to take when Compute Engine preempts the instance, either STOP (default) or DELETE.

     Important: Make sure your application can handle preemption. For example, handle preemption by specifying a shutdown script during instance creation. Learn how to handle preemption with a shutdown script.

2. Optional: If you chose to use a compact placement policy, then add the following flag to the command:

   --resource-policies=POLICY_NAME
   

   Replace the following:

   • POLICY_NAME: the name of the compact placement policy.

3. Optional: To attach Local SSD disks, then add the following --local-ssd flag for each Local SSD disk.

   Ensure that the number of Local SSD disks that you attach is allowed for your machine type; for details, see About Local SSD disks in the Compute Engine documentation. Specifically, G4 machine types use Titanium SSD disks.

   --local-ssd interface=INTERFACE_TYPE
   

   Replace the following:

   • INTERFACE_TYPE: the disk interface type that you want to use for the Local SSD disk. If your boot disk image has optimized NVMe drivers, then specify NVME. Specify SCSI for other images.

   After creating a VM with Local SSD disks, you must format and mount each device before you can use it.

4. Optional: To use NVIDIA RTX Virtual Workstation (vWS), add the following flag.

   When you create an instance that uses NVIDIA RTX Virtual Workstation, Compute Engine automatically adds a vWS license. For information about pricing for virtual workstations, see the GPU pricing page.

   --accelerator=type=nvidia-rtx-pro-6000-vws,count=VWS_ACCELERATOR_COUNT
   

   Replace VWS_ACCELERATOR_COUNT with the number of virtual GPUs that you want.

5. Run the command.

REST

To create instances in bulk, make a POST request to the instances.bulkInsert method.

The parameters that you need to specify depend on the consumption option that you are using for this deployment. Select the tab that corresponds to your consumption option.

Reservation

G4 instances support the following types of reservations:

• On-demand reservations
• Standard future reservations

To consume these types of reservations, an instance must use the standard provisioning model. For example, use the following creation parameters.

Before you submit the request, review all of the following steps to determine whether to include additional fields.

POST https://compute.googleapis.com/compute/v1/projects/PROJECT_ID/zones/ZONE/instances/bulkInsert
{
  "namePattern": "NAME_PATTERN",
  "count": "COUNT",
  "instanceProperties": {
    "machineType": "MACHINE_TYPE",
    "disks": [
      {
        "boot": true,
        "initializeParams": {
          "diskSizeGb": "DISK_SIZE",
          "diskType": "hyperdisk-balanced",
          "sourceImage": "projects/IMAGE_PROJECT/global/images/family/IMAGE_FAMILY"
        },
        "mode": "READ_WRITE",
        "type": "PERSISTENT"
      }
    ],
    "serviceAccounts": [
      {
        "email": "default",
        "scopes": [
          "https://www.googleapis.com/auth/cloud-platform"
        ]
      }
    ],
    "scheduling": {
    "provisioningModel": "STANDARD",
    "onHostMaintenance": "TERMINATE",
    "automaticRestart": AUTOMATIC_RESTART
  },
  "reservationAffinity": {
    "consumeReservationType": "RESERVATION_AFFINITY",
    "key": "compute.googleapis.com/reservation-name",
    "values": [
      "RESERVATION"
    ]
  }
  }
}

Complete the following steps:

1. Replace the following:

   • PROJECT_ID: the project ID of the project where you want to create the G4 instance.
   • ZONE: specify a zone in which the machine type that you want to use is available. If you want to specify a compact placement policy, then you must use a zone in the same region as the compact placement policy. For information about regions, see GPU availability by regions and zones.
   • NAME_PATTERN: the name pattern to use for the G4 instances. For example, using instance-# for the name pattern generates G4 instances with names such as instance-1 and instance-2, up to the number of G4 instances specified by --count.
   • COUNT: the number of G4 instances to create.
   • MACHINE_TYPE: the machine type to use for the G4 instance. For more information, see GPU machine types in the Compute Engine documentation.
   • DISK_SIZE: the size of the boot disk in GB. For more information, see either Google Cloud Hyperdisk size limits or Persistent Disk size limits based on the type of disk.
   • IMAGE_PROJECT: the project ID of the OS image.
   • IMAGE_FAMILY: the image family of the OS image that you want to use. For a list of all supported operating systems, see Supported operating systems.

   • AUTOMATIC_RESTART: if your instance automatically restarts if it crashes or Compute Engine stops it for a programmed stop, such as a maintenance event. Specify either true to enable automatic restart (default) or false to disable automatic restart.

   • RESERVATION_AFFINITY: which reservations the instance is allowed to consume, if possible. Specify one of the following:

     • ANY_RESERVATION (Reservation capacity is optional): this is the default option. The instance tries to use capacity from an automatically consumed reservation, which only happens if matching reservation capacity is available. Otherwise, the instance uses on-demand capacity.

     Caution: When matching reservation capacity is unavailable, the instance uses the on-demand consumption option instead, which has different billing and capacity assurance. For more information, see Choose a consumption option.

     • SPECIFIC_RESERVATION (Reservation capacity is required): Require the instance to use capacity from a specifically targeted reservation. The instance isn't allowed to use on-demand capacity. If reservation capacity is unavailable, then the instance creation fails.

   • RESERVATION: the reservation to use. Specify one of the following based on the value you specified for RESERVATION_AFFINITY:

     • If you specified ANY_RESERVATION, then delete RESERVATION so that the value is an empty string (""). Alternatively, you can remove the key and values fields entirely.

     • If you specified SPECIFIC_RESERVATION, then replace RESERVATION with the following:

       projects/RESERVATION_OWNER_PROJECT_ID/reservations/RESERVATION_NAME

       Replace the following:

       • RESERVATION_OWNER_PROJECT_ID: the project ID of the project where the reservation was created.
       Tip: If the reservation exists in the current project, then you can omit projects/RESERVATION_OWNER_PROJECT_ID/reservations/ from the RESERVATION value.
       • RESERVATION_NAME: the name of the reservation.

2. Optional: If you chose to use a compact placement policy, then add the following instanceProperties subfield to the request body:

       "resourcePolicies": [
         "projects/PROJECT_ID/regions/REGION/resourcePolicies/POLICY_NAME"
       ]
   

   Replace the following:

   • PROJECT_ID: the project ID of the compact placement policy.
   • REGION: the region of the compact placement policy.
   • POLICY_NAME: the name of the compact placement policy.

3. Optional: To use Local SSD disks, add the following field to the disks[] field for each Local SSD disk. Separate each disks[] subfield (including the boot disk) with a comma.

   Ensure that the number of Local SSD disks that you attach is allowed for your machine type; for details, see About Local SSD disks in the Compute Engine documentation. Specifically, G4 machine types use Titanium SSD disks.

       {
          "type": "SCRATCH",
          "initializeParams": {
             "diskType": "zones/ZONE/diskTypes/local-ssd"
          },
          "autoDelete": true,
          "interface": "INTERFACE_TYPE"
       }
   

   Replace the following:

   • ZONE: your specified zone.
   • INTERFACE_TYPE: the disk interface type that you want to use for the Local SSD disk. If your boot disk image has optimized NVMe drivers, then specify NVME. Specify SCSI for other images.

   After creating a VM with Local SSD disks, you must format and mount each device before you can use it.

4. Optional: To use NVIDIA RTX Virtual Workstation (vWS), add the following field.

   When you create an instance that uses NVIDIA RTX Virtual Workstation, Compute Engine automatically adds a vWS license. For information about pricing for virtual workstations, see the GPU pricing page.

   "guestAccelerators":
     [
       {
         "acceleratorCount": VWS_ACCELERATOR_COUNT,
         "acceleratorType": "projects/PROJECT_ID/zones/ZONE/acceleratorTypes/nvidia-rtx-pro-6000-vws"
       }
     ]
   

   Replace VWS_ACCELERATOR_COUNT with the number of virtual GPUs that you want.

5. Submit the request.

On-demand

Before you submit the request, review all of the following steps to determine whether to include additional fields.

POST https://compute.googleapis.com/compute/v1/projects/PROJECT_ID/zones/ZONE/instances/bulkInsert
{
  "namePattern": "NAME_PATTERN",
  "count": "COUNT",
  "instanceProperties": {
    "machineType": "MACHINE_TYPE",
    "disks": [
      {
        "boot": true,
        "initializeParams": {
          "diskSizeGb": "DISK_SIZE",
          "diskType": "hyperdisk-balanced",
          "sourceImage": "projects/IMAGE_PROJECT/global/images/family/IMAGE_FAMILY"
        },
        "mode": "READ_WRITE",
        "type": "PERSISTENT"
      }
    ],
    "serviceAccounts": [
      {
        "email": "default",
        "scopes": [
          "https://www.googleapis.com/auth/cloud-platform"
        ]
      }
    ],
    "scheduling": {
    "provisioningModel": "STANDARD",
    "onHostMaintenance": "TERMINATE",
    "automaticRestart": AUTOMATIC_RESTART
  },
  "reservationAffinity": {
    "consumeReservationType": "RESERVATION_AFFINITY"
  }
  }
}

Complete the following steps:

1. Replace the following:

   • PROJECT_ID: the project ID of the project where you want to create the G4 instance.
   • ZONE: specify a zone in which the machine type that you want to use is available. If you want to specify a compact placement policy, then you must use a zone in the same region as the compact placement policy. For information about regions, see GPU availability by regions and zones.
   • NAME_PATTERN: the name pattern to use for the G4 instances. For example, using instance-# for the name pattern generates G4 instances with names such as instance-1 and instance-2, up to the number of G4 instances specified by --count.
   • COUNT: the number of G4 instances to create.
   • MACHINE_TYPE: the machine type to use for the G4 instance. For more information, see GPU machine types in the Compute Engine documentation.
   • DISK_SIZE: the size of the boot disk in GB. For more information, see either Google Cloud Hyperdisk size limits or Persistent Disk size limits based on the type of disk.
   • IMAGE_PROJECT: the project ID of the OS image.
   • IMAGE_FAMILY: the image family of the OS image that you want to use. For a list of all supported operating systems, see Supported operating systems.

   • AUTOMATIC_RESTART: if your instance automatically restarts if it crashes or Compute Engine stops it for a programmed stop, such as a maintenance event. Specify either true to enable automatic restart (default) or false to disable automatic restart.

   • RESERVATION_AFFINITY: which reservations the instance is allowed to consume, if possible. Specify one of the following:

     • ANY_RESERVATION (Reservation capacity is optional): this is the default option. The instance tries to use capacity from an automatically consumed reservation, which only happens if matching reservation capacity is available. Otherwise, the instance uses on-demand capacity.

     Note: If you want using reservation capacity to be required, then follow the Reservation steps instead.

     • NO_RESERVATION (On-demand capacity is required): Block the instance from using any reservation capacity.

2. Optional: If you chose to use a compact placement policy, then add the following instanceProperties subfield to the request body:

       "resourcePolicies": [
         "projects/PROJECT_ID/regions/REGION/resourcePolicies/POLICY_NAME"
       ]
   

   Replace the following:

   • PROJECT_ID: the project ID of the compact placement policy.
   • REGION: the region of the compact placement policy.
   • POLICY_NAME: the name of the compact placement policy.

3. Optional: To use Local SSD disks, add the following field to the disks[] field for each Local SSD disk. Separate each disks[] subfield (including the boot disk) with a comma.

   Ensure that the number of Local SSD disks that you attach is allowed for your machine type; for details, see About Local SSD disks in the Compute Engine documentation. Specifically, G4 machine types use Titanium SSD disks.

       {
          "type": "SCRATCH",
          "initializeParams": {
             "diskType": "zones/ZONE/diskTypes/local-ssd"
          },
          "autoDelete": true,
          "interface": "INTERFACE_TYPE"
       }
   

   Replace the following:

   • ZONE: your specified zone.
   • INTERFACE_TYPE: the disk interface type that you want to use for the Local SSD disk. If your boot disk image has optimized NVMe drivers, then specify NVME. Specify SCSI for other images.

   After creating a VM with Local SSD disks, you must format and mount each device before you can use it.

4. Optional: To use NVIDIA RTX Virtual Workstation (vWS), add the following field.

   When you create an instance that uses NVIDIA RTX Virtual Workstation, Compute Engine automatically adds a vWS license. For information about pricing for virtual workstations, see the GPU pricing page.

   "guestAccelerators":
     [
       {
         "acceleratorCount": VWS_ACCELERATOR_COUNT,
         "acceleratorType": "projects/PROJECT_ID/zones/ZONE/acceleratorTypes/nvidia-rtx-pro-6000-vws"
       }
     ]
   

   Replace VWS_ACCELERATOR_COUNT with the number of virtual GPUs that you want.

5. Submit the request.

Spot

Before you submit the request, review all of the following steps to determine whether to include additional fields.

POST https://compute.googleapis.com/compute/v1/projects/PROJECT_ID/zones/ZONE/instances/bulkInsert
{
  "namePattern": "NAME_PATTERN",
  "count": "COUNT",
  "instanceProperties": {
    "machineType": "MACHINE_TYPE",
    "disks": [
      {
        "boot": true,
        "initializeParams": {
          "diskSizeGb": "DISK_SIZE",
          "diskType": "hyperdisk-balanced",
          "sourceImage": "projects/IMAGE_PROJECT/global/images/family/IMAGE_FAMILY"
        },
        "mode": "READ_WRITE",
        "type": "PERSISTENT"
      }
    ],
    "serviceAccounts": [
      {
        "email": "default",
        "scopes": [
          "https://www.googleapis.com/auth/cloud-platform"
        ]
      }
    ],
    "scheduling":
    {
      "provisioningModel": "SPOT",
      "instanceTerminationAction": "TERMINATION_ACTION",
      "onHostMaintenance": "TERMINATE",
      "automaticRestart": false
    }
  }
}

Complete the following steps:

1. Replace the following:

   • PROJECT_ID: the project ID of the project where you want to create the G4 instance.
   • ZONE: specify a zone in which the machine type that you want to use is available. If you want to specify a compact placement policy, then you must use a zone in the same region as the compact placement policy. For information about regions, see GPU availability by regions and zones.
   • NAME_PATTERN: the name pattern to use for the G4 instances. For example, using instance-# for the name pattern generates G4 instances with names such as instance-1 and instance-2, up to the number of G4 instances specified by --count.
   • COUNT: the number of G4 instances to create.
   • MACHINE_TYPE: the machine type to use for the G4 instance. For more information, see GPU machine types in the Compute Engine documentation.
   • DISK_SIZE: the size of the boot disk in GB. For more information, see either Google Cloud Hyperdisk size limits or Persistent Disk size limits based on the type of disk.
   • IMAGE_PROJECT: the project ID of the OS image.
   • IMAGE_FAMILY: the image family of the OS image that you want to use. For a list of all supported operating systems, see Supported operating systems.

   • TERMINATION_ACTION: the action to take when Compute Engine preempts the instance, either STOP (default) or DELETE.

     Important: Make sure your application can handle preemption. For example, handle preemption by specifying a shutdown script during instance creation. Learn how to handle preemption with a shutdown script.

2. Optional: If you chose to use a compact placement policy, then add the following instanceProperties subfield to the request body:

       "resourcePolicies": [
         "projects/PROJECT_ID/regions/REGION/resourcePolicies/POLICY_NAME"
       ]
   

   Replace the following:

   • PROJECT_ID: the project ID of the compact placement policy.
   • REGION: the region of the compact placement policy.
   • POLICY_NAME: the name of the compact placement policy.

3. Optional: To use Local SSD disks, add the following field to the disks[] field for each Local SSD disk. Separate each disks[] subfield (including the boot disk) with a comma.

   Ensure that the number of Local SSD disks that you attach is allowed for your machine type; for details, see About Local SSD disks in the Compute Engine documentation. Specifically, G4 machine types use Titanium SSD disks.

       {
          "type": "SCRATCH",
          "initializeParams": {
             "diskType": "zones/ZONE/diskTypes/local-ssd"
          },
          "autoDelete": true,
          "interface": "INTERFACE_TYPE"
       }
   

   Replace the following:

   • ZONE: your specified zone.
   • INTERFACE_TYPE: the disk interface type that you want to use for the Local SSD disk. If your boot disk image has optimized NVMe drivers, then specify NVME. Specify SCSI for other images.

   After creating a VM with Local SSD disks, you must format and mount each device before you can use it.

4. Optional: To use NVIDIA RTX Virtual Workstation (vWS), add the following field.

   When you create an instance that uses NVIDIA RTX Virtual Workstation, Compute Engine automatically adds a vWS license. For information about pricing for virtual workstations, see the GPU pricing page.

   "guestAccelerators":
     [
       {
         "acceleratorCount": VWS_ACCELERATOR_COUNT,
         "acceleratorType": "projects/PROJECT_ID/zones/ZONE/acceleratorTypes/nvidia-rtx-pro-6000-vws"
       }
     ]
   

   Replace VWS_ACCELERATOR_COUNT with the number of virtual GPUs that you want.

5. Submit the request.

For more information about the configuration options when creating VMs in bulk, see Create VMs in bulk in the Compute Engine documentation.

Install GPU drivers

After you create an instance, the instance can't use its GPUs unless the correct GPU drivers are already installed. The driver that you need to install depends on whether the instance has fractional GPUs or enabled NVIDIA RTX Virtual Workstation (vWS). Use one of the following options:

• If you specified a machine type with fractional GPUs, such as g4-standard-6, g4-standard-12, or g4-standard-24, then see Install vGPU drivers (fractional GPUs) in the Compute Engine documentation.
• If you enabled vWS, then see Install drivers for NVIDIA RTX Virtual Workstations (vWS) in the Compute Engine documentation.

• Otherwise, use the default GPU drivers as follows:

  • If you specified an OS image that includes GPU drivers, then no action is required.
  • If the OS image doesn't include GPU drivers, then see Install GPU drivers in the Compute Engine documentation.

Optional: Enable NVIDIA Multi-Instance GPU mode

Multi-Instance GPU (MIG) mode is a feature that you can enable for a supported NVIDIA GPU. After you create an instance, you can enable MIG mode on a single GPU that is attached to your machine. With MIG mode enabled, the single GPU is partitioned into as many as seven independent GPU instances. Each instance runs simultaneously, each with its own memory, cache, and streaming multiprocessors. You can then run different workloads on these GPU instances in parallel. MIG mode is distinct from using G4 machine types that have fractional GPUs attached, where multiple workloads share access to a single physical GPU through fractional (vGPUs) partitioning. For more information about using MIG mode, see the Multi-Instance GPU (MIG) User Guide in the NVIDIA documentation.

What's next

• View the details of a VM
• Get a list of VMs
• Verify reservation consumption
• Troubleshoot reservation consumption