N1 machine series

TheN1 general-purpose machine series supports several GPU models, and workstation configurations that specify any of the N1 machine types can also specify one of the following GPU models. For the chosen GPU model, the configuration may specify a count of how many GPU cards to attach to each workstation.

A2 machine series

TheA2 standard accelerator-optimized machine series has a fixed number of NVIDIA A100 GPUs attached, based solely on the chosen machine type.

The following table shows the mapping from machine type to the number of cards that will be attached.

A3 machine series

TheA3 Mega and A3 High accelerator-optimized machine series has a fixed number of NVIDIA H100 GPUs attached, based solely on the chosen machine type.

The following table shows the mapping from machine type to the number of cards that will be attached.

Console

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

gcloud

Install the Google Cloud CLI. After installation,initialize the Google Cloud CLI by running the following command:

gcloudinit

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

Console

Configure GPUs on an existing workstation configuration from theGoogle Cloud console, by doing the following:

1. In the Google Cloud console, go to theWorkstation configurations page.

   Go to Workstation configurations

2. In theWorkstation configurations list, click theName of theconfiguration to add GPUs to.

3. On theWorkstation configuration details page, clickeditEdit.

4. On theEdit workstation configuration page, clickMachine settingsin the navigation menu.

5. Click the toggle to selectGPUs instead of theGeneral purposemachine family.

   In theGPU type field, select the GPU Model that you want to use.

   In theNumber of GPUs field, select the number of GPU cards you wantto attach to each workstation.

   In theMachine type field, select the machine type you want to use.

   Note: the options for each field vary depending on thereplica zones chosen for the configuration and theGPU andmachine type availability inthose zones. If you cannot find suitable GPUs in the replica zones forthis configuration, you may decide tocreate a new configuration with GPUsinstead.

6. ClickSave to update the configuration.

gcloud

Configure GPUs on an existing workstation configuration by running thegcloud workstations configs update command.

First though, collect some information to see which GPU models areavailable and to choose one for your configuration:

1. Check which replica zones the configuration specifies by running thefollowinggcloud CLI command:

   gcloudworkstationsconfigsdescribe\--format="table(name.scope(workstationConfigs),replicaZones.list())"\--project=PROJECT_ID\--region=LOCATION\--cluster=CLUSTER_NAME\CONFIG_NAME

   Replace the following:

   • PROJECT_ID: the ID of the project that contains the workstation configuration.
   • LOCATION: the location of the workstation cluster.
   • CLUSTER_NAME: the name workstation cluster that contains the workstation configuration.
   • CONFIG_NAME: the name of the workstation configuration.

2. Choose a supported GPU model that is available inboth of theconfiguration's replica zones by running thegcloud compute accelerator-types list command:

   gcloudcomputeaccelerator-typeslist\--format="table(name:sort=1,zone,description,maximumCardsPerInstance)"\--filter='zone.basename()=(ZONES) AND name~"nvidia-tesla-(a100|p100|p4|t4|v100)$"'\--project=PROJECT_ID

   ReplaceZONES with a comma-separated list of thereplica zones determined in the previous step(for example,us-central1-a,us-central1-c).

   Choose a GPU model that is listed twice in the table, indicating it isavailable in both replica zones.

   Take note of the maximum cards you can attach for the chosen GPU model.

   Note: If there isn't suitableGPU Availability in thereplica zones for this configuration, you may decide tocreate a new configuration with GPUsinstead.

3. Determine which of the supported machine types are available inbothof the configuration's replica zones using thegcloud compute machine-types list command.

   1. If you chose the NVIDIA A100 40GB GPU model in the previous step,your configuration must use the A2 machine series:

      gcloudcomputemachine-typeslist\--format="table(name,zone,guestCpus:sort=1)"\--filter="name:a2-highgpu- OR name:a2-megagpu-"\--zones=ZONES\--project=PROJECT_ID

   2. If you chose any other GPU model in the previous step, yourconfiguration must use the N1 machine series:

      gcloudcomputemachine-typeslist\--format="table(name,zone,guestCpus:sort=1)"\--filter="name:n1-standard-"\--zones=ZONES\--project=PROJECT_ID

   Choose a machine type that islisted twice in the table, indicating itis available inboth replica zones.

4. Now that you have selected a GPU model and compatible machine type,update the configuration:

   1. For NVIDIA A100 40GB GPUs, run this command to update yourconfiguration:

      gcloudbetaworkstationsconfigsupdate\--project=PROJECT_ID\--region=LOCATION\--cluster=CLUSTER_NAME\CONFIG_NAME\--machine-type=A2_MACHINE_TYPE

      ReplaceA2_MACHINE_TYPE with the chosen A2 machinetype determined in the previous step (for example,a2-highgpu-1g).

      Note: for the A2 machine series the model and quantity of GPUs isfixed based on the machine type, so you may omit the--accelerator-type and--accelerator-count flags.

   2. For all other GPU models, run this command:

      gcloudbetaworkstationsconfigsupdate\--project=PROJECT_ID\--region=LOCATION\--cluster=CLUSTER_NAME\CONFIG_NAME\--machine-type=N1_MACHINE_TYPE\--accelerator-type=ACCELERATOR_TYPE\--accelerator-count=ACCELERATOR_COUNT

      Replace the following:

      • N1_MACHINE_TYPE: the chosen machine type from theN1 series (for example,n1-standard-2).
      • ACCELERATOR_TYPE: the chosen GPU model's name(for example,nvidia-tesla-t4).
      • ACCELERATOR_COUNT: the number of GPUs to attachto each workstation (for example,1,2,4). Must be a power of twoless than the maximum for the GPU model.

Install the Google Cloud CLI. After installation,initialize the Google Cloud CLI by running the following command:

gcloudinit

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

REST

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

Install the Google Cloud CLI. After installation,initialize the Google Cloud CLI by running the following command:

gcloudinit

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

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

Console

Create a new workstation configuration with GPUs from theGoogle Cloud console, by doing the following:

1. In the Google Cloud console, go to theWorkstation configurationspage.

   Go to Workstation configurations

2. On theWorkstation configurations page, clickadd_boxCreate.

3. On theBasic information step of theCreate workstation configuration page, specify a name for yourconfiguration in theName field.

   In theWorkstation cluster field, select a cluster in the chosenregion.

   Tip: GPUs are also attached toQuick start workstations. You can keepcosts down by selectingDisabled.

   ClickContinue to advance to theMachine settings step.

4. On theMachine settings step of theCreate workstationconfiguration page, begin by clicking the toggle to selectGPUsinstead of theGeneral purpose machine family.

   1. Then, in theZones field, select the checkboxes next to two zoneswhere the GPU model you chose is available (see theGPU availability table).

   2. In theGPU type field, select the GPU Model that you want to use.

   3. In theNumber of GPUs field, select the number of GPU cards youwant to attach to each workstation.

   4. In theMachine type field, select a compatible machine type.

5. Optional: This feature is not yet supported in the UI. See thegcloud instructions to use the CLI to configure this feature.

6. ClickContinue to configure the Environment settings and Identity and Access Management (IAM) policybefore clickingCreate to provision the new workstationconfiguration.

gcloud

Create a new workstation configuration with GPUs using thegcloud CLI, by running thegcloud workstations configs create command.

• For NVIDIA A100 40GB GPUs run this command to create your configuration:

  gcloudbetaworkstationsconfigscreate\--project=PROJECT_ID\--region=LOCATION\--cluster=CLUSTER_NAME\CONFIG_NAME\--replica-zones=REPLICA_ZONES\--machine-type=A2_MACHINE_TYPE

  Replace the following:

  • PROJECT_ID: the ID of the project that will contain the new workstation configuration.
  • LOCATION: the location of the workstation cluster where the configuration will be contained.
  • CLUSTER_NAME: the name workstation cluster that will contain the new workstation configuration.
  • CONFIG_NAME: the name of the new workstation configuration.
  • REPLICA_ZONES: exactly two zones in the cluster's region where the chosen GPU model is available (for example,us-central1-a,us-central1-c).
  • A2_MACHINE_TYPE: the chosen A2 series machine type (for example,a2-highgpu-1g).
  Note: for the A2 machine series the model and quantity of GPUs isfixed based on the machine type, so you may omit the--accelerator-type and--accelerator-count flags.

• For all other GPU models, run this command to create your configuration:

  gcloudbetaworkstationsconfigscreate\--project=PROJECT_ID\--region=LOCATION\--cluster=CLUSTER_NAME\CONFIG_NAME\--replica-zones=REPLICA_ZONES\--machine-type=N1_MACHINE_TYPE\--accelerator-type=ACCELERATOR_TYPE\--accelerator-count=ACCELERATOR_COUNT

  Replace the following:

  • PROJECT_ID: the ID of the project that will contain the new workstation configuration.
  • LOCATION: the location of the workstation cluster where the configuration will be contained.
  • CLUSTER_NAME: the name workstation cluster that will contain the new workstation configuration.
  • CONFIG_NAME: the name of the new workstation configuration.
  • REPLICA_ZONES: exactly two zones in the cluster's region where the chosen GPU model is available (for example,us-central1-a,us-central1-c).
  • N1_MACHINE_TYPE: the chosen N1 series machine type (for example,n1-standard-2).
  • ACCELERATOR_TYPE: the chosen GPU model's name(for example,nvidia-tesla-t4).
  • ACCELERATOR_COUNT: the number of GPUs to attachto each workstation (for example,1,2,4).

• Optional: To use a Hyperdisk, add flag--disk-type=hyperdisk-balanced-ha. For example:

  gcloudbetaworkstationsconfigscreate\...\--disk-type=hyperdisk-balanced-ha\--disk-size=200

  This command will create a Hyperdisk with size 200 GB. The disk type cannot be changed after creating the Workstation configuration.