Architecture

The following diagram shows the components used in this tutorial todeploy a single virtual workstation. Optional components shown in the diagraminclude different ways to connect to your virtual workstation, sharedstorage, an additional instance for serving third-party licenses, and additionalinfrastructure representing a render or compute farm.

Choose a region

An important factor when you deploy a virtual workstation is the latency betweenyour location and the instance you create. The lower the latency, the better theexperience. Therefore, you want to work in a region that's geographically closestto you. To learn more about where the different GPUs are available, seeGPU regions and zones availability.

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 region and zone that's physically closest to you.

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

   gcloud config set compute/zoneZONE

   ReplaceZONE with the name of thezoneyou're using, such asus-west1-b.

Choose a machine type

NVIDIA L4 GPUs are attached to theG2 machine type.Virtual machines with one or more GPUs have a maximum number of vCPUs for eachGPU you add to the instance. For example, each NVIDIA L4 GPU lets you haveup to 32 vCPUs and up to 128 GB of memory in your instance machine type. To seethe available vCPU and memory ranges for different GPU configurations, see theGPUs list.

The example in this tutorial consists of an 8 vCPU G2 virtual workstation, which iswell under the limit of 32 vCPUs for a single NVIDIA L4 GPU.

Create the virtual workstation

Note: There are somerestrictions to keep in mind when creating a virtual workstation with attached GPUs.Restrictions that apply to this tutorial have been addressed, but it's a goodidea to familiarize yourself with these restrictions before you create yourvirtual workstation.

1. In Cloud Shell, create the Compute Enginevirtual workstation instance:

   gcloudcomputeinstancescreateVM_NAME\--zone=ZONE\--machine-type=MACHINE_TYPE\--accelerator=type=ACCELERATOR,count=NUM-GPUS\--maintenance-policy="TERMINATE"\--image-project=ubuntu-os-cloud\--image-family=ubuntu-2204-lts\--boot-disk-size=SIZE\--boot-disk-type=TYPE\--network=NETWORK

   Replace the following:

   • VM_NAME is the name of your workstation.
   • ZONE is the zone in which to create the VM.
   • MACHINE_TYPE is the predefined or custommachine configuration.
   • ACCELERATOR is the type of GPU you want toattach, such asnvidia-l4-vws.
   • NUM-GPUS is the number of GPUs to attach to theVM.
   • SIZE is the size of the boot disk, in gigabytes.
   • TYPE is the type of boot disk. To get a list of available disk types, rungcloud compute disk-types list.
   • NETWORK is the network in which to create the VM.

   For example:

   gcloud compute instances create test-workstation \    --zone=us-west1-b \    --machine-type=g2-standard-8 \    --accelerator=type=nvidia-l4-vws,count=1 \    --maintenance-policy="TERMINATE" \    --image-project=ubuntu-os-cloud \    --image-family=ubuntu-2204-lts \    --boot-disk-size=100 \    --boot-disk-type=pd-ssd \    --network=default

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

   NAME: test-workstationZONE: us-west1-bMACHINE_TYPE: g2-standard-8PREEMPTIBLE:INTERNAL_IP: 10.138.XX.XXXEXTERNAL_IP: XX.XXX.XXX.XXXSTATUS: RUNNING

   Note the virtual workstation's external IP address. You will use it later inthe tutorial.

   You can retrieve the external IP address of your virtual workstation at anytime using theGoogle Cloud console.

Sign in to your virtual workstation

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

1. In Cloud Shell, connect to the new virtual workstation:

   gcloud compute sshVM_NAME

2. Set your account password. Logging into a virtual workstation using the PCoIPsoftware client requires a user password.

   sudo passwd `whoami`

   When you're prompted, enter a password. You will use this password laterin the tutorial to log on to your virtual workstation with the HP AnywarePCoIP Client.

Install the base libraries

The default Google Cloud Ubuntu 22.04 image is a minimal installation ofUbuntu. The next step is to install libraries that are necessary torun your virtual workstation as a graphics workstation.

1. Update the software repositories:

   sudo apt update

2. Install the base components:

   sudo apt install -y build-essentialsudo apt install -y libvulkan1

3. Update thegcc version for the NVIDIA driver:

   sudo apt install -y gcc-12sudo apt install -y linux-headers-$(uname -r)sudo update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-12 12sudo update-alternatives --config gcc

Install the NVIDIA driver

NVIDIA L4, NVIDIA T4, NVIDIA P4, and NVIDIA P100 GPUs work on Google Cloudonly with qualified NVIDIA RTX Virtual Workstation drivers for visualizationworkloads. These drivers can be downloaded from a public Google Cloud storage bucket.

1. To install the latest NVIDIA RTX Virtual Workstation driver, follow theinstructions (Steps 3 and 4, only).

2. Reboot the workstation:

   sudo reboot

Install the desktop environment

A desktop environment is necessary to run your virtual workstation as a graphicsworkstation. For this tutorial, you install the KDE Plasma Desktop.

1. Install the desktop environment:

   sudo apt updatesudo apt -y install kubuntu-desktopsudo apt -y install dialog

2. Reboot the workstation:

   sudo reboot

Install HP Anyware software

HP Anyware software provides a graphics agent that runs on yourvirtual workstation, delivering the desktop to your hardware or softwareclient.

Note: Installation may differ if you choose a different Linux distribution.For more information, seeHP Anyware Graphics Agent for Linux Installation Guide.

1. When the virtual workstation has restarted, in Cloud Shell,reconnect to the virtual workstation:

   gcloud compute sshVM_NAME

2. Add the Teradici software repository:

   curl -1sLf \    https://dl.anyware.hp.com/TOKEN/pcoip-agent/cfg/setup/bash.deb.sh \    | sudo -E distro=ubuntu codename=jammy bash

   Replace the following:

   • TOKEN is the download token you can retrieve fromtheHP Anyware Graphics Agent for Linuxpage underDownloads and scripts.

3. Update the software repositories:

   sudo apt update

4. Optional: Install USB dependencies, if you need to support USB devices otherthan keyboards, mice, and pointer devices.

   Note: If you skip this step, USB redirection is disabled. That means youcan't use bridged USB devices.

   sudo apt -y install usb-vhci-dkms

5. Install the HP Anyware software:

   sudo apt -y install pcoip-agent-graphics

Register the Anyware Graphics Agent

To use the Anyware Graphics Agent, you must have an HP Anyware license.

1. In Cloud Shell, activate your HP Anyware software license:

   pcoip-register-host --registration-code=REGISTRATION-CODE

   ReplaceREGISTRATION-CODE withthe code provided to youby HP Teradici in the formABCDEFGHIJKL@0123-4567-89AB-CDEF.

2. Reboot the virtual workstation:

   sudo reboot

Create a firewall rule

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

• In Cloud Shell (not on the virtual workstation), create a firewallrule that opens the required ports:

  gcloud compute firewall-rules create allow-pcoip \    --action=ALLOW \    --rules=tcp:443,tcp:4172,udp:4172 \    --source-ranges=0.0.0.0/0

  Note: This firewall rule allows traffic from anywhere on the internet toaccess your VM instance. Always followbest practices for firewall rules to restricttraffic to your instance.

Sign in to your virtual workstation using the PCoIP client

1. On your local computer, go to thePCoIP Clients sectionon theHP Anyware support page,and then download, install, and launchthe PCoIP Software Client application for your operating system.

2. In theHost Address or Registration Code field, enter the externalIP address of your virtual workstation. If you want, you can enter a name forthe connection.

   Note: By default, PCoIP traffic is encrypted using AES-256.However, HP Anyware Software uses a self-signed certificatethat's issued by Teradici. This might trigger a verification warning when youconnect. You can remove this warning by creating and installing your owncustom security certificates.Or, you can suppress the warning by changing theclient security modefor your client OS. For the purposes of this tutorial, you can ignore this warning by clickingConnect Insecurely in the dialog.

3. When you are connected, authenticate by entering the username andpassword that you created earlier for the virtual workstation.

   In a few seconds, your Linux desktop appears.

Test your virtual workstation

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

• Run GPU benchmark tools, such asglxgears,glmark2,orUNIGINE,which are programs to test graphics performance on a Linux workstation.
• InstallUnreal Engine,Unity Editor,Blender,or any content creation application.
• Run render benchmarking tools for popular renderers such asV-Ray,Octane,orMaxon.
• UseGoogle Chrometo browse your favorite sites or playYouTube videos.

You can also learn more aboutPCoIP performance optimizationbased on your workload.

gcloud compute instances stopVM_NAME