Creating a virtual GPU-accelerated Windows workstation

Stay organized with collections Save and categorize content based on your preferences.

This tutorial shows how to create a virtual Windows workstation running Windows Server 2016 with an attached display-capable GPU. (To create a Linux workstation, see the tutorial for Creating a virtual GPU-accelerated Linux workstation.) Google Cloud offers three display-capable GPUs: NVIDIA T4, NVIDIA Tesla P4, and NVIDIA Tesla P100.

After you create the virtual workstation, you learn how to remotely access it using Teradici PC-over-IP (PCoIP), a remote desktop protocol (RDP) widely used in the media and entertainment industry. PCoIP offers features essential to workloads media production such as color accuracy and support for lossless display.


  • Create a Compute Engine instance with a GPU. This instance serves as the foundation for a virtual workstation.
  • Install NVIDIA drivers on the virtual workstation.
  • Install Teradici Cloud Access Software on the virtual workstation.
  • Connect to the virtual workstation using a PCoIP software client or Zero Client, a type of hardware endpoint.


This tutorial uses the following billable components of Google Cloud:

You can use the pricing calculator to generate a cost estimate based on your projected usage. As of the time of this writing, the approximate cost for a typical media workstation configuration illustrated in this tutorial is US$2.26 per hour.

The resources that make up the virtual workstation and the factors that affect cost in this tutorial are:

Internet egress represents data that streams from your virtual workstation to your local display client and is billed at internet egress rates. Variables that affect data egress during a PCoIP session are bandwidth, screen resolution, number of display monitors, applications used, and the type of activity on each monitor. The cost in the example is based on an average usage of 10 Mbps. Teradici's Workflow Planning Guide can help you understand different workload requirements.

Before you begin

This tutorial uses gcloud and gsutil commands, which you can run from a Cloud Shell instance launched from the Google Cloud console . If you want to use gcloud and gsutil on your local workstation, install the Google Cloud CLI. The tutorial shows you how to run commands in Cloud Shell; if you use the gcloud CLI on your workstation, adjust the instructions accordingly.

  1. Sign in to your Google Cloud account. If you're new to Google Cloud, create an account to evaluate how our products perform in real-world scenarios. New customers also get $300 in free credits to run, test, and deploy workloads.
  2. In the Google Cloud console, on the project selector page, select or create a Google Cloud project.

    Go to project selector

  3. Make sure that billing is enabled for your Cloud project. Learn how to check if billing is enabled on a project.

  4. Enable the Compute Engine API.

    Enable the API

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

    Go to project selector

  6. Make sure that billing is enabled for your Cloud project. Learn how to check if billing is enabled on a project.

  7. Enable the Compute Engine API.

    Enable the API

In addition, make sure you have the following:

  • A Google Cloud project with quota for virtual workstation GPUs in your selected zone. You can get a list of GPU availability by using the gcloud compute accelerator-types list command.
  • A Google Chrome browser to access the Google Cloud console.
  • An RDP client of your choice. For more information, see Microsoft Remote Desktop clients.
  • A Teradici Zero Client or the latest Teradici software client for Windows, Mac, or Linux to access the virtual workstation.
  • A Teradici Cloud Access Software license. You can sign up for a trial license, or contact your Teradici representative. You will be provided with a 30-day trial registration code to use for this virtual workstation.

Understanding the architecture

The following diagram shows the components that are used in this tutorial to deploy a single virtual workstation. Optional components shown in the diagram include different ways to connect to your virtual workstation, object and shared storage, and an additional instance for serving third-party licenses.

Virtual workstation architecture.

Choosing an accelerator

An important factor when you deploy a virtual workstation is the latency between your location and the instance you create. The lower the latency, the better the experience. Therefore, you want to work in a zone that's geographically closest to you. To learn more about regions and zones, see Geography and regions.

  1. Open Cloud Shell. (If you're using the gcloud CLI, open a terminal window on your computer.)

    Go to Cloud Shell

  2. Get a list of the zones in which GPUs are available:

    gcloud compute accelerator-types list

    Take note of the zone that's physically closest to you.

  3. Set the zone that you want to work with:

    gcloud config set compute/zone zone

    Replace zone with the name of the zone you're using, such as us-west2-b.

Choosing a machine type

You can attach P4 GPUs to any machine type, but each GPU that's added to a virtual workstation must be coupled with a minimum number of vCPUs. This prevents oversubscription of any one resource. For the NVIDIA Tesla P4, you can attach up to 24 vCPUs to 1 GPU. To add more than 24 vCPUs to your virtual workstation, you must add an additional P4 GPU for every additional 24 vCPUs. For example, a 48-vCPU virtual workstation requires you to attach 2 P4 GPUs.

The example in this tutorial consists of a 24-vCPU virtual workstation, which is the maximum number of vCPUs allowed per NVIDIA Tesla P4 GPU.

Creating the virtual workstation

Teradici Graphics Agent (which you install on your virtual workstation later in this tutorial) requires you to enable IP forwarding and to allow HTTPS server traffic during virtual workstation creation.

  1. In Cloud Shell, create the Compute Engine virtual workstation instance. You must provide values for the placeholders such as name.

    gcloud compute instances create name \
        --machine-type machine-type \
        --accelerator type=accelerator,count=num-gpus \
        --can-ip-forward \
        --maintenance-policy "TERMINATE" \
        --tags "https-server" \
        --image-project windows-cloud \
        --image-family windows-2016 \
        --boot-disk-size size

    For example:

    gcloud compute instances create test-vws \
        --machine-type custom-24-32768 \
        --accelerator type=nvidia-tesla-p4-vws,count=1 \
        --can-ip-forward \
        --maintenance-policy "TERMINATE" \
        --tags "https-server" \
        --image-project windows-cloud \
        --image-family windows-2016 \
        --boot-disk-size 100

    After the virtual workstation is created, the machine status is displayed. The output is similar to the following:

    Created [].

    The virtual workstation is created in your project's default VPC network. If you want to create your virtual workstation in a different VPC network, add the following flag to the command:


    Replace network with the name of the network to use.

  2. Make a note of the virtual workstation's external IP address in the listing. You will use it later in the tutorial.

    You can retrieve the external IP address of your virtual workstation at any time from the Google Cloud console.

Creating a Windows password

You use RDP to access the Windows virtual workstation in order to install NVIDIA graphics drivers. To sign in with RDP, you need to create a Windows password for your user.

  1. Go to the Compute Engine section of the Google Cloud console.

    Go to the Compute Engine page

  2. Under the Connect column of your virtual workstation instance, select Set Windows password in the drop-down list.

    Setting the Windows password in the Google Cloud console.

    You are prompted to verify your username, and you see a warning about loss of access to encrypted data by resetting your password. You can ignore this warning, because this is a new disk and because there is no user data on your virtual workstation.

  3. When your Windows password is displayed, copy the password and then click Close.

    Make a note of this password, because you'll use it in the next step to sign in to your virtual workstation using the RDP client. Keep this password in a secure location.

Signing in to your virtual workstation

After you create the virtual workstation, sign in to the machine so that you can configure it.

  1. In the Compute Engine section of the Google Cloud console, click the RDP dropdown and select the Download the RDP file option to download the RDP file for your instance. Use this file to connect to the instance using an RDP client. For more information, see Microsoft Remote Desktop clients.

The virtual workstation desktop appears. Wait for Windows Server Manager to finish launching before you continue to the next step. When Server Manager is ready, you see a display like the following:

Windows Server Manager displayed in the virtual workstation.

Configuring IE Enhanced Security

The default Windows Server 2016 image has default security settings that prevent you from downloading files from the internet. To download the NVIDIA driver, you need to disable these security features.

  1. In the left-hand side of Server Manager, click Local Server.
  2. Next to IE Enhanced Security Configuration, click On.
  3. In the dialog, turn off Enhanced Security for both Administrators and Users.

    Enhanced Security dialog.

  4. Close both the IE Enhanced Security Configuration and Server Manager windows.

Installing the NVIDIA driver

NVIDIA T4, NVIDIA Tesla P4, and NVIDIA Tesla P100 GPUs work on Google Cloud only with qualified NVIDIA Quadro Virtual Data Center Workstation (vWS) drivers for both compute and display workloads. You can download these drivers from a public storage bucket.

  1. On the Windows desktop of your virtual workstation, open the Cloud Shell by double-clicking its icon.

  2. Get a listing of the latest drivers:

    gsutil ls gs://nvidia-drivers-us-public/GRID

    For this tutorial, you use the latest graphics drivers that are available at the time of writing:

    GRID11.1 Windows driver (452.39_grid_win10_server2016_server2019_64bit_international.exe)

    You can always find the latest qualified driver under the GRID directory that has the highest version number; if you find a newer driver, use that one.

  3. In the Google gcloud CLI Shell, download and install the driver. If you're using a version of the driver that's more recent than 452.39, change the command accordingly.

    gsutil -m cp gs://nvidia-drivers-us-public/GRID/GRID11.1/452.39_grid_win10_server2016_server2019_64bit_international.exe %USERPROFILE%/Downloads
  4. Open Windows File Explorer and go to the Downloads directory for your user.

  5. In the directory, right-click the NVIDIA driver in the directory and then select Run as administrator.

    Follow the prompts to install the driver in the default location:

    1. If you accept the terms of the NVIDIA license, click Agree and Continue.
    2. Select Express (Recommended) as the installation type.
    3. When the installation is finished, close the NVIDIA installer.
  6. Verify that the driver is working. In the Cloud Shell, run this command (including the double quotation marks):

    "C:\Program Files\NVIDIA Corporation\NVSMI\nvidia-smi.exe"

    The output is similar to the following:

    | NVIDIA-SMI 452.39       Driver Version: 452.39       CUDA Version: 11.0     |
    | GPU  Name            TCC/WDDM | Bus-Id        Disp.A | Volatile Uncorr. ECC |
    | Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
    |                               |                      |               MIG M. |
    |   0  Tesla P4           WDDM  | 00000000:00:04.0 Off |                    0 |
    | N/A   37C    P8     8W /  75W |    222MiB /  7680MiB |      7%      Default |
    |                               |                      |                  N/A |
    | Processes:                                                                  |
    |  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
    |        ID   ID                                                   Usage      |
    |    0   N/A  N/A         8    C+G   ...dows\System32\LogonUI.exe    N/A      |
    |    0   N/A  N/A      6756    C+G   C:\Windows\System32\dwm.exe     N/A      |
    |    0   N/A  N/A      7544    C+G   C:\Windows\System32\dwm.exe     N/A      |

    If you don't see output like what's in this listing, refer to the Troubleshooting section later in this tutorial.

Installing Teradici Cloud Access Software

Teradici Cloud Access Software provides a graphics agent that runs on your virtual workstation, delivering the desktop to your hardware or software client.

  1. On the Windows instance, open Internet Explorer and go to the PCoIP Hosts section on the Teradici support page.
  2. Download Teradici Graphics Agent for Windows.
  3. Complete the agent installation process and choose the option to run the downloaded graphics agent executable. Follow the prompts to install the graphics agent in the default location.

Registering Teradici Graphics Agent

When the installation process is complete, you're prompted for a registration code.

  1. Enter the registration code you got earlier.

    Registering the Teradici registration code.

  2. After you've entered the registration code, reboot the Windows instance.

Creating a firewall rule

The PCoIP client communicates with your virtual workstation using several ports. You must set firewall rules that allow traffic to and from your virtual workstation.

  • In Cloud Shell on your local computer (not on the virtual workstation), create a firewall rule that opens the required ports:

    gcloud compute firewall-rules create allow-teradici \
        --allow tcp:443,tcp:4172,udp:4172,tcp:60443

Signing in to your virtual workstation using the PCoIP client

  1. On your local computer, go to the PCoIP Clients section on the Teradici support page, and then download, install, and launch the PCoIP Client application for your operating system.

  2. Select New Connection.

  3. In the Host Address field, enter the external IP address of your virtual workstation. If you want, you can enter a name for the connection.

    Assigning an external IP address to the virtual workstation.

  4. When you are connected, authenticate by entering the username and password that you created earlier for the virtual workstation.

    Entering the username and password for accessing the virtual workstation.

  5. Select the desktop to run and then click Connect.

    Connecting to the virtual workstation.

    In a few seconds, you see your Windows desktop.

    Windows desktop showing in the virtual workstation.

Testing your virtual workstation

After you've deployed your virtual workstation, you can test performance and interactivity using a number of tools:

  • Use one of the NVIDIA GPU demos to test graphics performance on a Windows workstation.
  • Install Blender, an open source 3D software package.
  • Run render benchmarking tools for popular renderers such as V-Ray, Octane, or Maxon.
  • Install Google Chrome to browse your favorite sites or play YouTube videos.


This section lists issues you might encounter when you set up or connect to the workstation.

Audio does not work

Issue: You do not hear any sound from your Windows virtual workstation.

Solution: Because of how Teradici Graphics Agent works, you might need to manually enable the Windows Audio Service on your instance.

  1. Sign in to the Windows instance, and then click Start > Control Panel > Hardware > Sound to open the Sound Control Panel.

    Error message showing Windows Audio Service not enabled.

  2. Click Yes. Windows verifies that the audio driver is working.

    Windows Sound dialog showing Teradici virtual audio driver.

    If sound is working, you hear sound from your Windows virtual workstation.

NVIDIA-SMI has failed

Issue: NVIDIA-SMI has failed because it couldn't communicate with the NVIDIA driver.

Solution: Reinstall the driver and try running nvidia-smi again.

Unable to connect to the virtual workstation

Issue: You are using a PCoIP Zero Client, and you're unable to connect to your virtual workstation.

Solution: Ensure that your Zero Client has firmware version 6.1 or later installed before you connect to the virtual workstation. For more information, contact your Teradici representative.

Clean up

To avoid incurring charges to your Google Cloud account for the resources used in this tutorial, either delete the project that contains the resources, or keep the project and delete the individual resources.

After you've finished the tutorial, clean up the resources you created on Google Cloud so you won't be billed for them in the future.

Stopping your virtual workstation

Stopped virtual workstations incur costs for persistent disk usage, but they can be restarted at any time. To stop your virtual workstation, run the following command in Cloud Shell:

gcloud compute instances stop test-vws

Delete the project

  1. In the Google Cloud console, go to the Manage resources page.

    Go to Manage resources

  2. In the project list, select the project that you want to delete, and then click Delete.
  3. In the dialog, type the project ID, and then click Shut down to delete the project.

Delete all the components

  1. Delete the virtual workstation.
  2. Delete the firewall rule.

What's next