
Advantages of GPU virtualization
In a virtualized environment that is powered by NVIDIA graphics cards, the NVIDIA virtual GPU (vGPU) software is installed on the virtualization layer along with the hypervisor. This software creates virtual GPUs that allow each virtual machine (VM) to share the physical GPU installed on the server. For more demanding calculations, one virtual machine can use the performance of several physical GPUs at the same time.
NVIDIA vGPU software includes a graphics or compute driver for each virtual machine. As the processes that used to be calculated on the CPU are transferred to the GPU, users get much better performance. In virtualized and cloud environments, GPUs can handle demanding engineering and graphics applications, as well as computationally intensive tasks such as AI and data science.
Unlimited GPU performance
The performance of virtualized GPUs is not degraded and reaches almost the same values as physical GPU cards.
Easy to set up
Using common data center management tools, management and monitoring of virtualized GPUs is easier and more secure.
Optimal use of resources
Combine the performance of several GPU cards, or on the contrary, divide them into several smaller instances, exactly according to your needs.
Change performance as needed
Easily add, remove or change profiles as your graphics performance needs evolve.
VIRTUAL GPU licenses
NVIDIA Virtual GPU (vGPU) offers a wide range of uses – from virtual desktops and high-performance virtual workstations to virtual servers for data science and AI. NVIDIA vGPU enables the combination of all the advantages of a virtualized environment – security, simple management and full performance of NVIDIA graphics cards.
NVIDIA vGPU software, whether installed in the cloud or on a data center server, creates virtual GPUs that can be shared between multiple virtual machines. Any device can connect to them remotely – from the office, from home, simply from anywhere.
NVIDIA Virtual Applications (vApps)
Increase productivity and performance with GPU virtualization with NVIDIA Virtual Applications (vApps). The vApps software is suitable for accelerating application streaming with Remote Desktop Sharing Host (RDSH) solutions, including Citrix Virtual Apps and VMware Horizon vApps. With GPU sharing, multiple virtual machines can be powered by a single GPU, increasing efficiency and affordability. NVIDIA vApps enables users to access any Windows application at full power, on any device and most importantly from anywhere.
NVIDIA Virtual PC (vPC)
NVIDIA vPC and NVIDIA graphics cards can accelerate most desktop applications, and thanks to them, users can easily access all the tools they need from anywhere. Whether you’re at home or in the office, the way you work is changing dynamically, and today’s applications demand exponentially higher graphics performance. You can use it not only with tools like MS Teams and Zoom, which help teams collaborate in real time regardless of location, but also in situations where employees need several monitors for their work.
NVIDIA RTX Virtual Workstation (vWS)
NVIDIA vWS, combined with powerful NVIDIA graphics processors, is a dynamic force for visual computing, enabling you to run the most powerful virtual workstations from the data center or the cloud on any device. Graphic artists, developers, 3D designers or engineers can connect to the most demanding applications from anywhere and, above all, with high performance that rivals physical workstations. And all this at lower costs and higher security.
NVIDIA Virtual Compute Server (vCS)
NVIDIA vCS enables data centers running Red Hat Enterprise Linux, Red Hat Virtualization, and other supported KVM-based hypervisors to accelerate virtualized server computing using the latest NVIDIA A100 and A30 accelerators. Computationally intensive tasks such as artificial intelligence training, deep learning, and data science can be run on these virtual servers. In conjunction with NVIDIA AI Enterprise, you get a professional solution for AI.
Features and usage of VIRTUAL GPU licenses
vApps | vPC | vWS | vCS | |
Usual use | Virtualization and remote access to a specific application | Desktop virtualization for regular users | Powerful workstation virtualization for graphic designers and engineers | A virtual computing server for AI and data analysis |
Suggested GPU | A2, A16 | A2, A16 | A10, A40 | A30, A100 |
Licensed per | Concurent Users (CCU) | Concurent Users (CCU) | Concurent Users (CCU) | GPU |
Max memory per profile | 32 GB | 2 GB | 48 GB | 80 GB |
Max profiles per GPU | 32 | 32 | 32 | 10 |
Max displays / Max resolution | 1 1280 × 1024 | 4 5120 × 2880 (5K) | 4 7680 × 4320 (8K) | 1 4096 × 2160 (4K) |
Multi-vGPU | ➖ | ➖ | ✔️ | ✔️ |
Virt. desktop | ➖ | ✔️ | ✔️ | ➖ |
Compute accelleration | ➖ | ➖ | ✔️ | ✔️ |
Supports OS Windows | ✔️ | ✔️ | ✔️ | ➖ |
Supports OS Linux | ➖ | ✔️ | ✔️ | ✔️ |
NVIDIA Graphics Driver | ✔️ | ✔️ | ✔️ | ➖ |
CUDA, OpenCL | ➖ | ➖ | ✔️ | ✔️ |
OpenGL, DirectX, Vulkan | ✔️ | ✔️ | ✔️ | ➖ |
M Computers represents NVIDIA on the Czech market in the Enterprise area of computing accelerators and AI systems.
As the first company in Central and Eastern Europe, we received the highest ELITE PARTNER status and at the same time two awards NVIDIA AI Innovator and NVIDIA AI Champion.
Testing
Not sure which GPU card or license type is right for you? To test performance and, above all, deployment speed, we have NVIDIA A100, A30, A40, A16 and A2 accelerators and all types of demo vGPU licenses available. If you are interested in our testing offer, please fill out this form.