VDI/RDSH Desktops

Q: I have 500 users of whom 300 are connected to VDI desktops concurrently. How many and what types of NVIDIA virtual GPU (vGPU) software licenses do I need?

A: 300 GRID vPC or Quadro vDWS licenses are needed. Each virtual desktop requires a license after it is started and running. One GRID vPC or Quadro vDWS license is needed for each concurrent VM.

Q: I use Citrix Virtual Desktop or VMware Horizon for VDI which also has apps published within it from a Virtual Apps, Horizon Apps, or RDSH server. Which NVIDIA vGPU software license do I need?

A: One GRID vPC or Quadro vDWS license for each VDI desktop is needed. Both NVIDIA GRID Virtual PC and Quadro vDWS include a per-user entitlement for GRID vApps.

Q: What happens if my virtual desktop cannot acquire an NVIDIA vGPU license from the license server?

A: For the first 20 minutes, the vGPU or physical GPU assigned to the VM operates at full capacity. When licensing is enforced through software, the performance of the virtual GPU or physical GPU is degraded over time if the VM fails to obtain a license.

Q: I want to use GPU pass through or Hyper-V DDA with an NVIDIA GPU that supports NVIDIA vGPU for RDSH-based desktops. What license do I need?

A: One NVIDIA GRID vApps license is needed for each concurrent user session.

Q: I want to pass-through a card that vGPU software does not support (for example, . Quadro P4000 or GeForce 2080 Ti) to a virtual machine and use it with RDSH-based published desktops. Do I need a license?

A: NVIDIA does not support or recommend such a configuration for commercial or education use.

Q: I want to run an NVIDIA GPU in a bare-metal deployment with Windows Server 2016/2019 installed directly on the hardware (without a hypervisor). I then need to publish user desktops or apps on this server using RDSH, Citrix Virtual Apps, or VMware Horizon. Which NVIDIA vGPU software license do I need?

A: One NVIDIA GRID vApps license is required for each concurrent RDSH-based desktop or app user.

Note: NVIDIA vGPU software graphics drivers are also required where graphics acceleration is needed in bare meta deployments. These drivers are available only from the NVIDIA Licensing Portal. The drivers available from NVIDIA Driver Downloads are optimised for compute workloads or workstations, not virtual graphics acceleration.

Q: When would I require a Quadro Virtual Data Center Workstation (Quadro vDWS) license instead of a GRID vPC license?

A: Professional graphics users who require Quadro-level performance (for example, app certification, higher frame rates, CUDA support, rendering, or CAD) require one Quadro vDWS license per running VM. Knowledge workers using Microsoft Office and traditional office applications such as Adobe Acrobat and web browsers require only a GRID vPC license per running VM.

Q: What are some examples of applications for which NVIDIA recommends Quadro vDWS?

A: Autodesk AutoCAD, Autodesk Revit, Solidworks 3DExperience, Siemens NX, PTC Creo, Dassault Systèmes CATIA, Esri ArcGIS, Adobe Creative Cloud, Schlumberger Petrel, Ansys Discovery Live, medicalPACs apps, and RTX Rendering Apps.

Non VDI/RDSH Deployment Scenarios

Q: I have 1000 users that use Citrix Virtual Apps but only 600 of them are connected to a session at any given time. How many and what types of vGPU software licenses do I need?

A: 600 GRID vApps licenses are needed. One GRID vApps license is needed for each concurrent user.

Q: I have a Windows or Linux based application that uses an NVIDIA GPU but does not use a traditional remoting solution(such as RDS, Citrix, Horizon) to provide graphics to a user. Instead, it provides graphics output to users through a browser-based web front-end. How many and what types of vGPU software licenses do I need?

A: For most customer use cases, a GRID vApps license would be recommended for app streaming and session based solutions. For example, for applications such as Siemens TeamCenter, one GRID vApps license is required for each concurrent user.

If an application requires any of the Quadro vDWS features, you will need a Quadro vDWS license for each user.

Q: I want to run a vGPU supported NVIDIA GPU in a bare-metal deployment with Windows server 2016/2019 installed directly on the hardware (without a hypervisor). My application then uses the GPU to render graphics but does NOT use RDSH session-based desktops to display the results to a user. Which vGPU license do I need?

A: For most customer use cases, a GRID vApps license would be recommended for app streaming and session based solutions.

If an application requires any of the Quadro vDWS features, you will need a Quadro vDWS license for each user.

Note: NVIDIA vGPU software graphics drivers are also required where graphics acceleration is needed in bare metal deployments. These drivers are available only from the NVIDIA Software Licensing Portal. The drivers available from NVIDIA Driver Downloads are optimised for compute workloads or workstations, not virtual graphics acceleration.

General

Q: Do I need a license to run NVIDIA vGPU software?

A: Yes. Refer to the packaging, pricing, and licensing guide for detailed information on licensing.

Q: Can I use my NVIDIA vGPU licenses in the cloud?

A: Yes. NVIDIA vGPU software is supported on several cloud services with bring your own license (BYOL) licensing and licensing provided by the cloud service.

Q: How are NVIDIA GRID vApps licenses enforced?

A: How NVIDIA GRID vApps licenses are enforced depends on the type of deployment for which the license is being used:

Q: I have purchased a co-termed subscription license. Can I assign the licenses that have not reached the start date yet to my license server?

A: This practice is not recommended. If the licenses that have not reached their start date are assigned to your license server, the license server will show the active and inactive licenses even though the co-termed license is not active yet .

Q: I have purchased a co-termed subscription license. Will the license server start using the co-termed licenses that have not reached the start date yet if I remove the licenses that are still active?

A: No. The license server will only use licenses that are currently active. The active license must remain on the license server until the co-termed license reaches its start date.

Q: From where do I download the NVIDIA virtual GPU software?

A: NVIDIA virtual GPU software is available to customers from the NVIDIA Software Licensing Portal,which you can access by logging in to the NVIDIA Application Hub.

If you have not already purchased NVIDIA virtual GPU software and would like to try it, you can obtain a limited trial license for evaluation. Note that you will need a supported NVIDIA GPU to use the license.

Q: Is my server certified for NVIDIA vGPU software?

A: NVIDIA’s network of OEMs certify a wide range of server models. Check the NVIDIA Certified Servers site or visit your OEM’s website for details.

Q: Which hypervisors are supported with NVIDIA vGPU software?

A: See the product support matrix for a list of supported hypervisors. For detailed compatibility information, see the hypervisor software release notes on the NVIDIA vGPU software documentation website.

Q: Can I get an evaluation license?

A: Yes, evaluation licenses are available for 128 users for 90 days on all NVIDIA vGPU GPU software products.

Q: Which remoting solutions does NVIDIA vGPU software support?

A: See the product support matrix for the remoting solutions that are supported with each supported hypervisor.

Q: How do I get support for NVIDIA vGPU software?

A: You can get support from NVIDIA Virtual GPU Support Services.

You require a current Support, Upgrade and Maintenance (SUMs) contract to log a support call. Refer to your partner or NVIDIA account manager for presales or evaluation support enquiries.

Q: For how long is my release of NVIDIA vGPU software supported?

A: The support policy depends on whether your release is from a Long Term Service Branch (LTSB) or a New Feature Branch (NFB):

Support dates for each version are listed on the vGPU software documentation home page.