Create an A3 Ultra or A4 instance

This document describes how to create instances with attached GPUs from the A3 Ultra or A4 machine series. To learn more about creating instances with attached GPUs, see Overview of creating an instance with attached GPUs.

Before you begin

To review limitations and additional prerequisite steps for creating instance with attached GPUs, such as how to select an OS image or check GPU quota, see Overview of creating an instance with attached GPUs.
If you haven't already, then set up authentication. Authentication is the process by which your identity is verified for access to Google Cloud services and APIs. To run code or samples from a local development environment, you can authenticate to Compute Engine by selecting one of the following options:

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
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.
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 instances, 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 instances. To see the exact permissions that are required, expand the Required permissions section:

Required permissions

The following permissions are required to create instances:

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 assign a legacy network to the VM: compute.networks.use on the project
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 legacy network: compute.networks.useExternalIp on the project
To specify a subnet for your VM: compute.subnetworks.use on the project or on the chosen subnet
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 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.

Determine how to create A3 Ultra or A4 instances

A3 Ultra or A4 instances are available through the following creation options, which each have different creation procedures, resource availability, and pricing. Identify which option that you want to use based on your workload.

If you need to run long-running AI and ML workloads, such as large model training and inferencing, that require the lowest latency, then create VMs or clusters that use the features and services available from Cluster Director. With Cluster Director, you can reserve densely allocated machines that provide topology-aware scheduling, as well as enhanced monitoring and maintenance. To learn more about Cluster Director, see Cluster Director in the AI Hypercomputer documentation.

For instructions to create A3 Ultra or A4 instances, see Overview of creating VMs and clusters in the AI Hypercomputer documentation.
If you need to shorter-term duration AI and ML workloads, then choose one of the following options:
- If your workloads need to run for up to 90 days with minimal network latency, then reserve densely allocated machines for a future date by using future reservation requests in calendar mode. At the start of your reservation period, you can create A3 Ultra or A4 instances by using the reserved capacity. At the end of your reservation period, Compute Engine deletes the reservation and any VMs that consume it.
  
  For instructions to create A3 Ultra or A4 instances that use a reservation, see Create an A3 Ultra or A4 instance in this document.
- If your workloads can start at any time and need an exact number of VMs to start, then use resize requests in a managed instance group (MIG). With MIG resize requests, you ask Compute Engine to add VMs all at once to your MIG whenever resources are available. These VMs can run for up to seven days. Because resources are delivered from a secured pool, you have higher chances of obtaining GPUs than if you were to make an on-demand request.
  
  For more information about MIG resize requests, see About resize requests in a MIG.
- If your workloads tolerate availability disruptions, then you can get significant discounts by using Spot VMs. You can create and delete Spot VMs as needed. However, Spot VMs are finite resources and Compute Engine might preempt (automatically stop or delete) Spot VMs at any time to reclaim capacity. To learn more about Spot VMs, see Spot VMs.
  
  For instructions on how to create A3 Ultra or A4 instances by using Spot VMs, see Create an A3 Ultra or A4 instance in this document.

Create an A3 Ultra or A4 instance

To create an A3 Ultra or A4 instance by using a reservation or Spot VMs, complete the following steps:

Create VPC networks
Create the instance
Prepare the instance for use

Create VPC networks

Tip: If you are setting up a quick test, you can skip this step and specify a single NIC --network-interface=nic-type=GVNIC instead.

For A4 or A3 Ultra machine type, you must create three VPC networks for the following network interfaces:

2 VPC networks for the gVNIC network interfaces (NIC). These are used for host to host communication.
1 VPC network with the RDMA network profile is required for the CX7 NICs. This network needs to have 8 subnets, one subnet for each CX7 NIC, and is used for GPU to GPU communication.

For more information about NIC arrangement, see Review network bandwidth and NIC arrangement.

Set up the networks either manually by following the instruction guides or automatically by using the provided script.

Instruction guides

To create the networks, you can use the following instructions:

To create the VPC networks for the gVNICs, see Create and manage Virtual Private Cloud networks.
To create the VPC network with the RDMA network profile, see Create a Virtual Private Cloud network for RDMA NICs.

For these VPC networks, we recommend setting the maximum transmission unit (MTU) to a larger value. For A4 or A3 Ultra machine type, the recommended MTU is 8896 bytes. To review the recommended MTU settings for other GPU machine types, see MTU settings for GPU machine types.

Script

To create the networks, you can use the following script.

  #!/bin/bash

  # Create standard VPCs (network and subnets) for the gVNICs
  for N in $(seq 0 1); do
    gcloud compute networks create GVNIC_NAME_PREFIX-net-$N \
      --subnet-mode=custom \
      --mtu=8896

    gcloud compute networks subnets create GVNIC_NAME_PREFIX-sub-$N \
      --network=GVNIC_NAME_PREFIX-net-$N \
      --region=REGION \
      --range=10.$N.0.0/16

    gcloud compute firewall-rules create GVNIC_NAME_PREFIX-internal-$N \
      --network=GVNIC_NAME_PREFIX-net-$N \
      --action=ALLOW \
      --rules=tcp:0-65535,udp:0-65535,icmp \
      --source-ranges=10.0.0.0/8
  done

  # Create SSH firewall rules
  gcloud compute firewall-rules create GVNIC_NAME_PREFIX-ssh \
    --network=GVNIC_NAME_PREFIX-net-0 \
    --action=ALLOW \
    --rules=tcp:22 \
    --source-ranges=IP_RANGE

  # Assumes that an external IP is only created for vNIC 0
  gcloud compute firewall-rules create GVNIC_NAME_PREFIX-allow-ping-net-0 \
    --network=GVNIC_NAME_PREFIX-net-0 \
    --action=ALLOW \
    --rules=icmp \
    --source-ranges=IP_RANGE

  # List and make sure network profiles exist
  gcloud compute network-profiles list

  # Create network for CX-7
  gcloud compute networks create RDMA_NAME_PREFIX-mrdma \
    --network-profile=ZONE-vpc-roce \
    --subnet-mode custom \
    --mtu=8896

  # Create subnets.
  for N in $(seq 0 7); do
    gcloud compute networks subnets create RDMA_NAME_PREFIX-mrdma-sub-$N \
      --network=RDMA_NAME_PREFIX-mrdma \
      --region=REGION \
      --range=10.$((N+2)).0.0/16 # offset to avoid overlap with gVNICs
  done

Replace the following:

GVNIC_NAME_PREFIX: the custom name prefix to use for the standard VPC networks and subnets for the gVNICs.
RDMA_NAME_PREFIX: the custom name prefix to use for the VPC network and subnets with the RDMA network profile for the CX7 NICs.
ZONE: specify a zone in which the machine type that you want to use is available. For information about regions, see GPU availability by regions and zones.
REGION: the region where you want to create the subnets. This region must correspond to the zone specified. For example, if your zone is europe-west1-b, then your region is europe-west1.
IP_RANGE: the IP range to use for the SSH firewall rules.

Create the instance

To create an A3 Ultra or A4 instance, select one of the following methods:

Create an instance that uses a reservation
Create a Spot VM

Create an instance that uses a reservation

To create an instance that uses a reservation, use one of the following options:

Console

In the Google Cloud console, go to the Create an instance page.

Go to Create an instance

The Create an instance screen appears and displays the Machine configuration pane.
In the Machine configuration pane, complete the following steps:
1. Specify a Name for your instance. See Resource naming convention.
2. Select the Region and Zone where you have reserved capacity.
3. Click the GPUs tab, and then complete the following steps:
  1. In the GPU type list, select your GPU type.
    - For A4 instances, select NVIDIA B200.
    - For A3 Ultra instances, select NVIDIA H200 141GB.
  2. In the Number of GPUs list, select 8.
In the navigation menu, click OS and storage. In the OS and storage pane that appears, complete the following steps:
1. Click Change. The Boot disk configuration pane appears.
2. On the Public images tab, select a recommended image. For a list of recommended images, see Operating systems.
3. To confirm your boot disk options, click Select.
To create a multi-NIC instance, complete the following steps. Otherwise, to create a single-NIC instance, skip these steps.
- In the navigation menu, click Networking. In the Networking pane that appears, complete the following steps:
  1. In the Network interfaces section, complete the following steps:
  2. Delete the default network interface. To delete the interface, click Delete.
  3. Click Add a network interface. Use this option to add the gVNIC and RDMA networks that you created in the previous section. When you add the networks, remember the following:
    - Specify your host networks in the Network and Subnetwork lists, and set the Network interface card list to gVNIC.
    - Specify your GPU networks in the Network and Sub-network lists, and set the Network interface card list to MRDMA for these networks.
In the navigation menu, click Advanced. In the Advanced pane that appears, complete the following steps:
1. In the Reservations section, select Choose a reservation, and then click Choose reservation. This action opens a pane with a list of available reservations within your selected zone. From the reservation list, complete the following steps:
  1. Select the reservation that you want to use for the VM. You can also select a specific block within the reservation.
  2. Click Choose.
2. In the Provisioning model section, expand the VM provisioning model advanced settings section.
3. In the On VM termination list, select Stop (default) or Delete.
To create and start the instance, click Create.

gcloud

To create the VM, use the gcloud compute instances create command.

gcloud compute instances create VM_NAME  \
    --machine-type=MACHINE_TYPE \
    --image-family=IMAGE_FAMILY \
    --image-project=IMAGE_PROJECT \
    --zone=ZONE \
    --boot-disk-type=hyperdisk-balanced \
    --boot-disk-size=DISK_SIZE \
    --scopes=cloud-platform \
    --network-interface=nic-type=GVNIC,network=GVNIC_NAME_PREFIX-net-0,subnet=GVNIC_NAME_PREFIX-sub-0 \
    --network-interface=nic-type=GVNIC,network=GVNIC_NAME_PREFIX-net-1,subnet=GVNIC_NAME_PREFIX-sub-1,no-address \
    --network-interface=nic-type=MRDMA,network=RDMA_NAME_PREFIX-mrdma,subnet=RDMA_NAME_PREFIX-mrdma-sub-0,no-address \
    --network-interface=nic-type=MRDMA,network=RDMA_NAME_PREFIX-mrdma,subnet=RDMA_NAME_PREFIX-mrdma-sub-1,no-address \
    --network-interface=nic-type=MRDMA,network=RDMA_NAME_PREFIX-mrdma,subnet=RDMA_NAME_PREFIX-mrdma-sub-2,no-address \
    --network-interface=nic-type=MRDMA,network=RDMA_NAME_PREFIX-mrdma,subnet=RDMA_NAME_PREFIX-mrdma-sub-3,no-address \
    --network-interface=nic-type=MRDMA,network=RDMA_NAME_PREFIX-mrdma,subnet=RDMA_NAME_PREFIX-mrdma-sub-4,no-address \
    --network-interface=nic-type=MRDMA,network=RDMA_NAME_PREFIX-mrdma,subnet=RDMA_NAME_PREFIX-mrdma-sub-5,no-address \
    --network-interface=nic-type=MRDMA,network=RDMA_NAME_PREFIX-mrdma,subnet=RDMA_NAME_PREFIX-mrdma-sub-6,no-address \
    --network-interface=nic-type=MRDMA,network=RDMA_NAME_PREFIX-mrdma,subnet=RDMA_NAME_PREFIX-mrdma-sub-7,no-address \
    --reservation-affinity=specific \
    --reservation=RESERVATION \
    --provisioning-model=RESERVATION_BOUND \
    --instance-termination-action=TERMINATION_ACTION \
    --maintenance-policy=TERMINATE

Replace the following:

VM_NAME: the name of the VM.
MACHINE_TYPE: the machine type to use for the VM. Specify either an A4 or A3 Ultra machine type. For more information, see GPU machine types.
IMAGE_FAMILY: the image family of the OS image that you want to use. For a list of supported operating systems, see Operating system details.
IMAGE_PROJECT: the project ID of the OS image.
ZONE: the zone in which the machine type that you want to use is available. For information about regions, see GPU regions and zones.
DISK_SIZE: the size of the boot disk in GB.
GVNIC_NAME_PREFIX: the name prefix that you specified when creating the standard VPC networks and subnets that use gVNICs.
RDMA_NAME_PREFIX: the name prefix that you specified when creating the VPC networks and subnets that use RDMA NICs.

RESERVATION: either the reservation name or a specific block within a reservation. To get the reservation name or the available blocks, see View reserved capacity. Choose one of the following:

Reservation value When to use

Reservation value	When to use
`projects/RESERVATION_PROJECT_ID/reservations/RESERVATION_NAME`	Use this value when one of the following is true: *You are* using a compact placement policy. Compute Engine applies the policy to the reservation and creates the VMs on a single block. You aren't using a compact placement policy, and it's acceptable to create the VMs on any of the reserved blocks.**
`projects/RESERVATION_PROJECT_ID/reservations/RESERVATION_NAME/reservationBlocks/RESERVATION_BLOCK_NAME`	Use this value when *you aren't* using a compact placement policy, and you want to create VMs on a specific reserved block.**


projects/RESERVATION_PROJECT_ID/reservations/RESERVATION_NAME

Use this value when one of the following is true:

You are using a compact placement policy. Compute Engine applies the policy to the reservation and creates the VMs on a single block.
You aren't using a compact placement policy, and it's acceptable to create the VMs on any of the reserved blocks.


projects/RESERVATION_PROJECT_ID/reservations/RESERVATION_NAME/reservationBlocks/RESERVATION_BLOCK_NAME

Use this value when you aren't using a compact placement policy, and you want to create VMs on a specific reserved block.

Tip: If the reservation is located in the current project, you can omit projects/RESERVATION_PROJECT_ID/reservations/ from the reservation value.

TERMINATION_ACTION: whether Compute Engine stops (STOP) or deletes (DELETE) the VM at the end of the reservation period.

REST

To create the VM, make a POST request to the instances.insert method.

POST https://compute.googleapis.com/compute/v1/projects/PROJECT_ID/zones/ZONE/instances
{
  "machineType":"projects/PROJECT_ID/zones/ZONE/machineTypes/MACHINE_TYPE",
  "name":"VM_NAME",
  "disks":[
    {
      "boot":true,
      "initializeParams":{
        "diskSizeGb":"DISK_SIZE",
        "diskType":"hyperdisk-balanced",
        "sourceImage":"projects/IMAGE_PROJECT/global/images/family/IMAGE_FAMILY"
      },
      "mode":"READ_WRITE",
      "type":"PERSISTENT"
    }
  ],
  "networkInterfaces": [
    {
      "accessConfigs": [
        {
          "name": "external-nat",
          "type": "ONE_TO_ONE_NAT"
        }
      ],
      "network": "projects/NETWORK_PROJECT_ID/global/networks/GVNIC_NAME_PREFIX-net-0",
      "nicType": "GVNIC",
      "subnetwork": "projects/NETWORK_PROJECT_ID/region/REGION/subnetworks/GVNIC_NAME_PREFIX-sub-0"
    },
    {
      "network": "projects/NETWORK_PROJECT_ID/global/networks/GVNIC_NAME_PREFIX-net-1",
      "nicType": "GVNIC",
      "subnetwork": "projects/NETWORK_PROJECT_ID/region/REGION/subnetworks/GVNIC_NAME_PREFIX-sub-1"
    },
    {
      "network": "projects/NETWORK_PROJECT_ID/global/networks/RDMA_NAME_PREFIX-mrdma",
      "nicType": "MRDMA",
      "subnetwork": "projects/NETWORK_PROJECT_ID/region/REGION/subnetworks/RDMA_NAME_PREFIX-mrdma-sub-0"
    },
    {
      "network": "projects/NETWORK_PROJECT_ID/global/networks/RDMA_NAME_PREFIX-mrdma",
      "nicType": "MRDMA",
      "subnetwork": "projects/NETWORK_PROJECT_ID/region/REGION/subnetworks/RDMA_NAME_PREFIX-mrdma-sub-1"
    },
    {
      "network": "projects/NETWORK_PROJECT_ID/global/networks/RDMA_NAME_PREFIX-mrdma",
      "nicType": "MRDMA",
      "subnetwork": "projects/NETWORK_PROJECT_ID/region/REGION/subnetworks/RDMA_NAME_PREFIX-mrdma-sub-2"
    },
    {
      "network": "projects/NETWORK_PROJECT_ID/global/networks/RDMA_NAME_PREFIX-mrdma",
      "nicType": "MRDMA",
      "subnetwork": "projects/NETWORK_PROJECT_ID/region/REGION/subnetworks/RDMA_NAME_PREFIX-mrdma-sub-3"
    },
    {
      "network": "projects/NETWORK_PROJECT_ID/global/networks/RDMA_NAME_PREFIX-mrdma",
      "nicType": "MRDMA",
      "subnetwork": "projects/NETWORK_PROJECT_ID/region/REGION/subnetworks/RDMA_NAME_PREFIX-mrdma-sub-4"
    },
    {
      "network": "projects/NETWORK_PROJECT_ID/global/networks/RDMA_NAME_PREFIX-mrdma",
      "nicType": "MRDMA",
      "subnetwork": "projects/NETWORK_PROJECT_ID/region/REGION/subnetworks/RDMA_NAME_PREFIX-mrdma-sub-5"
    },
    {
      "network": "projects/NETWORK_PROJECT_ID/global/networks/RDMA_NAME_PREFIX-mrdma",
      "nicType": "MRDMA",
      "subnetwork": "projects/NETWORK_PROJECT_ID/region/REGION/subnetworks/RDMA_NAME_PREFIX-mrdma-sub-6"
    },
    {
      "network": "projects/NETWORK_PROJECT_ID/global/networks/RDMA_NAME_PREFIX-mrdma",
      "nicType": "MRDMA",
      "subnetwork": "projects/NETWORK_PROJECT_ID/region/REGION/subnetworks/RDMA_NAME_PREFIX-mrdma-sub-7"
    }
  ],
  "reservationAffinity":{
    "consumeReservationType":"SPECIFIC_RESERVATION",
    "key":"compute.googleapis.com/reservation-name",
    "values":[
      "RESERVATION"
    ]
  },
"scheduling":{
    "provisioningModel":"RESERVATION_BOUND",
    "instanceTerminationAction":"TERMINATION_ACTION",
    "onHostMaintenance": "TERMINATE",
    "automaticRestart":true
  }
}

Replace the following:

PROJECT_ID: the project ID of the project where you want to create the VM.
ZONE: the zone in which the machine type that you want to use is available. For information about regions, see GPU regions and zones.
MACHINE_TYPE: the machine type to use for the VM. Specify either an A4 or A3 Ultra machine type. For more information, see GPU machine types.
VM_NAME: the name of the VM.
DISK_SIZE: the size of the boot disk in GB.
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 supported operating systems, see Operating system details.
NETWORK_PROJECT_ID: the project ID of the network.
GVNIC_NAME_PREFIX: the name prefix that you specified when creating the standard VPC networks and subnets that use gVNICs.
REGION: the region of the subnetwork.
RDMA_NAME_PREFIX: the name prefix that you specified when creating the VPC networks and subnets that use RDMA NICs.

RESERVATION: either the reservation name or a specific block within a reservation. To get the reservation name or the available blocks, see View reserved capacity. Choose one of the following:

Reservation value When to use

Reservation value	When to use
`projects/RESERVATION_PROJECT_ID/reservations/RESERVATION_NAME`	Use this value when one of the following is true: *You are* using a compact placement policy. Compute Engine applies the policy to the reservation and creates the VMs on a single block. You aren't using a compact placement policy, and it's acceptable to create the VMs on any of the reserved blocks.**
`projects/RESERVATION_PROJECT_ID/reservations/RESERVATION_NAME/reservationBlocks/RESERVATION_BLOCK_NAME`	Use this value when *you aren't* using a compact placement policy, and you want to create VMs on a specific reserved block.**


projects/RESERVATION_PROJECT_ID/reservations/RESERVATION_NAME

Use this value when one of the following is true:

You are using a compact placement policy. Compute Engine applies the policy to the reservation and creates the VMs on a single block.
You aren't using a compact placement policy, and it's acceptable to create the VMs on any of the reserved blocks.


projects/RESERVATION_PROJECT_ID/reservations/RESERVATION_NAME/reservationBlocks/RESERVATION_BLOCK_NAME

Use this value when you aren't using a compact placement policy, and you want to create VMs on a specific reserved block.

Tip: If the reservation is located in the current project, you can omit projects/RESERVATION_PROJECT_ID/reservations/ from the reservation value.

TERMINATION_ACTION: whether Compute Engine stops (STOP) or deletes (DELETE) the VM at the end of the reservation period.

Create a Spot VM

To create the Spot VM, use one of the following options:

Console

In the Google Cloud console, go to the Create an instance page.

Go to Create an instance

The Create an instance screen appears and displays the Machine configuration pane.
In the Machine configuration pane, complete the following steps:
1. Specify a Name for your instance. See Resource naming convention.
2. Select the Region and Zone where you want to create a VM. To check in which regions and zones GPUs are available, see GPU regions and zones.
3. Click the GPUs tab, and then complete the following steps:
  1. In the GPU type list, select your GPU type.
    - For A4 instances, select NVIDIA B200.
    - For A3 Ultra instances, select NVIDIA H200 141GB.
  2. In the Number of GPUs list, select 8.
In the navigation menu, click OS and storage. In the OS and storage pane that appears, complete the following steps:
1. Click Change. The Boot disk configuration pane appears.
2. On the Public images tab, select a recommended image. For a list of recommended images, see Operating systems.
3. To confirm your boot disk options, click Select.
To create a multi-NIC instance, complete the following steps. Otherwise, to create a single-NIC instance, skip these steps.
- In the navigation menu, click Networking. In the Networking pane that appears, complete the following steps:
  1. In the Network interfaces section, complete the following steps:
  2. Delete the default network interface. To delete the interface, click Delete.
  3. Click Add a network interface. Use this option to add the gVNIC and RDMA networks that you created in the previous section. When you add the networks, you must do the following:
    - Specify your host networks in the Network and Subnetwork lists, and set the Network interface card list to gVNIC.
    - Specify your GPU networks in the Network and Sub-network lists, and set the Network interface card list to MRDMA for these networks.
In the navigation menu, click Advanced. In the Advanced pane that appears, complete the following steps:
1. In the Provisioning model section, select Spot in the VM provisioning model list.
2. Optional: To specify the action to take when Compute Engine preempts the instance, complete the following steps:
  1. Expand the VM provisioning model advanced settings section.
  2. In the On VM termination list, select Stop or Delete.
To create and start the instance, click Create.

gcloud

To create the VM, use the gcloud compute instances create command.

gcloud compute instances create VM_NAME  \
    --machine-type=MACHINE_TYPE \
    --image-family=IMAGE_FAMILY \
    --image-project=IMAGE_PROJECT \
    --zone=ZONE \
    --boot-disk-type=hyperdisk-balanced \
    --boot-disk-size=DISK_SIZE \
    --scopes=cloud-platform \
    --network-interface=nic-type=GVNIC,network=GVNIC_NAME_PREFIX-net-0,subnet=GVNIC_NAME_PREFIX-sub-0 \
    --network-interface=nic-type=GVNIC,network=GVNIC_NAME_PREFIX-net-1,subnet=GVNIC_NAME_PREFIX-sub-1,no-address \
    --network-interface=nic-type=MRDMA,network=RDMA_NAME_PREFIX-mrdma,subnet=RDMA_NAME_PREFIX-mrdma-sub-0,no-address \
    --network-interface=nic-type=MRDMA,network=RDMA_NAME_PREFIX-mrdma,subnet=RDMA_NAME_PREFIX-mrdma-sub-1,no-address \
    --network-interface=nic-type=MRDMA,network=RDMA_NAME_PREFIX-mrdma,subnet=RDMA_NAME_PREFIX-mrdma-sub-2,no-address \
    --network-interface=nic-type=MRDMA,network=RDMA_NAME_PREFIX-mrdma,subnet=RDMA_NAME_PREFIX-mrdma-sub-3,no-address \
    --network-interface=nic-type=MRDMA,network=RDMA_NAME_PREFIX-mrdma,subnet=RDMA_NAME_PREFIX-mrdma-sub-4,no-address \
    --network-interface=nic-type=MRDMA,network=RDMA_NAME_PREFIX-mrdma,subnet=RDMA_NAME_PREFIX-mrdma-sub-5,no-address \
    --network-interface=nic-type=MRDMA,network=RDMA_NAME_PREFIX-mrdma,subnet=RDMA_NAME_PREFIX-mrdma-sub-6,no-address \
    --network-interface=nic-type=MRDMA,network=RDMA_NAME_PREFIX-mrdma,subnet=RDMA_NAME_PREFIX-mrdma-sub-7,no-address \
    --provisioning-model=SPOT \
    --instance-termination-action=TERMINATION_ACTION

Replace the following:

VM_NAME: the name of the VM.
MACHINE_TYPE: the machine type to use for the VM. Specify either an A4 or A3 Ultra machine type. For more information, see GPU machine types.
IMAGE_FAMILY: the image family of the OS image that you want to use. For a list of supported operating systems, see Operating system details.
IMAGE_PROJECT: the project ID of the OS image.
ZONE: the zone in which the machine type that you want to use is available. For information about regions, see GPU regions and zones.
DISK_SIZE: the size of the boot disk in GB.
GVNIC_NAME_PREFIX: the name prefix that you specified when creating the standard VPC networks and subnets that use gVNICs.
RDMA_NAME_PREFIX: the name prefix that you specified when creating the VPC networks and subnets that use RDMA NICs.
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, we recommend that you handle preemption by specifying a shutdown script during instance creation. Learn how to handle preemption with a shutdown script.

REST

To create the VM, make a POST request to the instances.insert method.

POST https://compute.googleapis.com/compute/v1/projects/PROJECT_ID/zones/ZONE/instances
{
  "machineType":"projects/PROJECT_ID/zones/ZONE/machineTypes/MACHINE_TYPE",
  "name":"VM_NAME",
  "disks":[
    {
      "boot":true,
      "initializeParams":{
        "diskSizeGb":"DISK_SIZE",
        "diskType":"hyperdisk-balanced",
        "sourceImage":"projects/IMAGE_PROJECT/global/images/family/IMAGE_FAMILY"
      },
      "mode":"READ_WRITE",
      "type":"PERSISTENT"
    }
  ],
  "networkInterfaces": [
    {
      "accessConfigs": [
        {
          "name": "external-nat",
          "type": "ONE_TO_ONE_NAT"
        }
      ],
      "network": "projects/NETWORK_PROJECT_ID/global/networks/GVNIC_NAME_PREFIX-net-0",
      "nicType": "GVNIC",
      "subnetwork": "projects/NETWORK_PROJECT_ID/region/REGION/subnetworks/GVNIC_NAME_PREFIX-sub-0"
    },
    {
      "network": "projects/NETWORK_PROJECT_ID/global/networks/GVNIC_NAME_PREFIX-net-1",
      "nicType": "GVNIC",
      "subnetwork": "projects/NETWORK_PROJECT_ID/region/REGION/subnetworks/GVNIC_NAME_PREFIX-sub-1"
    },
    {
      "network": "projects/NETWORK_PROJECT_ID/global/networks/RDMA_NAME_PREFIX-mrdma",
      "nicType": "MRDMA",
      "subnetwork": "projects/NETWORK_PROJECT_ID/region/REGION/subnetworks/RDMA_NAME_PREFIX-mrdma-sub-0"
    },
    {
      "network": "projects/NETWORK_PROJECT_ID/global/networks/RDMA_NAME_PREFIX-mrdma",
      "nicType": "MRDMA",
      "subnetwork": "projects/NETWORK_PROJECT_ID/region/REGION/subnetworks/RDMA_NAME_PREFIX-mrdma-sub-1"
    },
    {
      "network": "projects/NETWORK_PROJECT_ID/global/networks/RDMA_NAME_PREFIX-mrdma",
      "nicType": "MRDMA",
      "subnetwork": "projects/NETWORK_PROJECT_ID/region/REGION/subnetworks/RDMA_NAME_PREFIX-mrdma-sub-2"
    },
    {
      "network": "projects/NETWORK_PROJECT_ID/global/networks/RDMA_NAME_PREFIX-mrdma",
      "nicType": "MRDMA",
      "subnetwork": "projects/NETWORK_PROJECT_ID/region/REGION/subnetworks/RDMA_NAME_PREFIX-mrdma-sub-3"
    },
    {
      "network": "projects/NETWORK_PROJECT_ID/global/networks/RDMA_NAME_PREFIX-mrdma",
      "nicType": "MRDMA",
      "subnetwork": "projects/NETWORK_PROJECT_ID/region/REGION/subnetworks/RDMA_NAME_PREFIX-mrdma-sub-4"
    },
    {
      "network": "projects/NETWORK_PROJECT_ID/global/networks/RDMA_NAME_PREFIX-mrdma",
      "nicType": "MRDMA",
      "subnetwork": "projects/NETWORK_PROJECT_ID/region/REGION/subnetworks/RDMA_NAME_PREFIX-mrdma-sub-5"
    },
    {
      "network": "projects/NETWORK_PROJECT_ID/global/networks/RDMA_NAME_PREFIX-mrdma",
      "nicType": "MRDMA",
      "subnetwork": "projects/NETWORK_PROJECT_ID/region/REGION/subnetworks/RDMA_NAME_PREFIX-mrdma-sub-6"
    },
    {
      "network": "projects/NETWORK_PROJECT_ID/global/networks/RDMA_NAME_PREFIX-mrdma",
      "nicType": "MRDMA",
      "subnetwork": "projects/NETWORK_PROJECT_ID/region/REGION/subnetworks/RDMA_NAME_PREFIX-mrdma-sub-7"
    }
  ],
  "scheduling":
  {
    "provisioningModel": "SPOT",
    "instanceTerminationAction": "TERMINATION_ACTION"
  }
}

Replace the following:

PROJECT_ID: the project ID of the project where you want to create the VM.
ZONE: the zone in which the machine type that you want to use is available. For information about regions, see GPU regions and zones.
MACHINE_TYPE: the machine type to use for the VM. Specify either an A4 or A3 Ultra machine type. For more information, see GPU machine types.
VM_NAME: the name of the VM.
DISK_SIZE: the size of the boot disk in GB.
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 supported operating systems, see Operating system details.
NETWORK_PROJECT_ID: the project ID of the network.
GVNIC_NAME_PREFIX: the name prefix that you specified when creating the standard VPC networks and subnets that use gVNICs.
REGION: the region of the subnetwork.
RDMA_NAME_PREFIX: the name prefix that you specified when creating the VPC networks and subnets that use RDMA NICs.
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, we recommend that you handle preemption by specifying a shutdown script during instance creation. Learn how to handle preemption with a shutdown script.

Prepare the instance for use

To prepare an instance that has GPUs attached for use, complete the following steps:

To enable an A4 or A3 Ultra instance to use its attached GPUs, the instance must have GPU drivers installed. Unless the image in the instance already includes the required GPU drivers, install GPU drivers.
If you created a Spot VM in the previous section, then complete the following steps:
- To prepare your Spot VM for a potential preemption, see Manage preemption of Spot VMs.
- Optional: Learn about best practices for Spot VMs.

What's next

To monitor GPU performance, see Monitor GPU performance.
To troubleshoot GPU instances, see Troubleshoot GPU VMs.
Learn more about GPU platforms.

Create an A3 Ultra or A4 instance Stay organized with collections Save and categorize content based on your preferences.

Before you begin

Console

gcloud

REST

Required roles

Required permissions

Determine how to create A3 Ultra or A4 instances

Create an A3 Ultra or A4 instance

Create VPC networks

Instruction guides

Script

Create the instance

Create an instance that uses a reservation

Console

gcloud

REST

Create a Spot VM

Console

gcloud

REST

Prepare the instance for use

What's next

Create an A3 Ultra or A4 instance