This section covers selecting the correct vGPU Profiles:
Creative and Technical Professionals
Knowledge Worker profiles
Frame buffer Utilization and vGPU Profile selection
Selecting the right vGPU profile based on the users’ tasks and workloads maximizes the whole virtualization experience. Virtual GPU profiles determine the amount of frame buffer allocated to your virtual machine and which license is used for the virtual instances. This section provides vGPU profile guidance and additional references for these NVIDIA licensed software products - NVIDIA RTX Virtual Workstation (RTX vWS), NVIDIA Virtual PC (vPC), and NVIDIA Virtual Applications (vApps).
Previously, creative and technical professionals were limited to physical devices like laptops or desktop workstations due to their demanding tasks and heavy workloads. But now, the NVIDIA Virtual GPU (vGPU) paired with the NVIDIA RTX Virtual Workstation (RTX vWS) software solution enables creative and technical professionals the ability to access their most demanding applications from anywhere with performance that rivals physical workstations. NVIDIA RTX vWS software accelerates professional design and visualization applications, including Autodesk Revit, Maya, Dassault Systèmes CATIA, Solidworks, Esri ArcGIS Pro, Petrel, and more.
Considerations for selecting the right vGPU profile for creative and technical professionals are compatibility and performance. The Q-series vGPU profiles undergo the same rigorous application certification process as the NVIDIA RTX™ Enterprise platform drivers or professional graphics applications. RTX vWS software supports RTX Enterprise drivers, allowing users to benefit from the acceleration and stability that RTX brings to professional applications used by the most demanding customers today. As a result, you can expect 100% compatibility and performance within your applications using the NVIDIA RTX vWS software licensed product.
Begin your profile selection by considering the requirements of your users’ primary applications. Professional application software vendors certify their products to run with NVIDIA vGPU software, ensuring that the performance is tuned for maximum efficiency. Most vendors’ websites have a dedicated page indicating the proper GPU hardware; use those recommendations to select the appropriate vGPU profile to meet your end users’ needs. To understand more about the graphics requirements of your users’ applications, consult the application vendors.
Matching Profiles to User Needs
As stated earlier in the Sizing Your Environment section, you must define your users’ needs and match them to the NVIDIA vGPU profiles that provide the right amount of frame buffer and the correct software licenses. By categorizing a user’s workload as Light, Medium, and Heavy, we can illustrate a base VM configuration for an NVIDIA RTX vWS deployment.
Light user
8 GB RAM
Four vCPUs (2.4 GHz)
A40-8Q vGPU Profile
Medium
16 GB RAM
Eight vCPUs (2.6 GHz)
A40-12Q
Heavy user
32 GB RAM
12 vCPUs (3.2 GHz)
A40-24Q
These VM configurations are benchmark testing for the RTX vWS “Dedicated Performance” user type quality of service (QoS). For more RTX vWS Benchmarking results and vGPU sizing guidance, refer to the NVIDIA RTX vWS Workstation Sizing Guide.
Today’s digital workplace has become a hybrid workforce that is more now than ever dependent on desktop virtualization with increased graphics requirements for productivity applications. The workload of the digital worker (aka Knowledge Worker) includes a graphics-rich experience with immersive visual quality and processing speed. Selecting the right NVIDIA virtual GPU (vGPU) profile can meet the increasing demands of the Knowledge worker in a virtual desktop (VDI) environment. NVIDIA vGPU profiles determine the amount of frame buffer allocated to your virtual machine. The vGPU profiles supported on NVIDIA GPUs with NVIDIA software are the 1B (with 1024 MB of frame buffer) and 2B (with 2048 MB of frame buffer).
The NVIDIA vPC licensed software is selected for knowledge worker workloads within a virtual desktop infrastructure (VDI) environment. Workloads configured with the NVIDIA vPC software accelerates office productivity applications, streaming video, Windows, RDSH, multiple and high-resolution monitors, and 2D electric design automation (EDA).
Matching Profiles to User Needs
As stated earlier in the Sizing Your Environment section, you must define your users’ needs and match them to the NVIDIA vGPU profiles that provide the right amount of resources. Here is a recommended knowledge worker’s virtual machine configuration.
vPC-1B vGPU profile for a Horizon VDI Desktop
6 GB RAM
Four vCPUs @ 2.4 GHz or faster (single-socket)
Dual HD Monitor - (1920x1080)
vPC-2B vGPU profile for a Horizon VDI Desktop
6 GB RAM
Four vCPUs @ 2.4 GHz or faster (single-socket)
Dual Quad Monitor (2560x1440) or Single 4K Monitor (4096x2160)
For more information on Knowledge Worker sizing guidance, refer to the NVIDIA vPC Sizing Guide.
Frame Buffer Utilization and vGPU Profile selection
As stated earlier, NVIDIA vGPU profiles determine the amount of frame buffer allocated to your virtual machine, thereby rendering frame buffer utilization a critical metric when considering vGPU Profiles for your VDI deployment. Frame buffer utilization within the VM can be affected by the application load, monitor configurations, and screen resolution.
Multiple monitors and higher screen resolutions scenarios are also impactful considerations when deciding your deployment’s most optimal vGPU profile. When the number of monitors is increased, more pixels are being delivered to the screen. The NVIDIA Engineering team conducted benchmark testing to illustrate the increased advantages of VDI desktops configured with NVIDIA’s vPC software over CPU-Only configured VDI desktops.