What is a vGPU? An Overview of Virtual GPUs
Virtual GPU (vGPU) technology enables multiple users or virtual machines to share a single physical GPU, allowing access without dedicating the entire card to individual instances. This approach is particularly effective when various workloads require GPU acceleration, yet assigning a physical GPU to every user would be inefficient.
What Is a vGPU?
A vGPU represents a specific portion of a physical GPU allocated to a virtual machine or user. By partitioning the physical hardware into dedicated slices, each user is granted their own isolated VRAM and GPU resources.
For instance, a single physical GPU can host several vGPUs. Each virtual machine perceives its assigned resources as if they were a full card, rather than seeing the underlying physical hardware. This setup permits simultaneous usage by multiple users on the same GPU.
It is important to note that a vGPU functions differently from simple GPU sharing between applications. Instead of sharing, the GPU resources are segmented and assigned to individual virtual machines.
How Does vGPU Work?
Once a physical GPU is installed in a host system, virtualization software and compatible GPU technology partition its resources into multiple virtual GPUs.
- Physical GPU: The host system retains the actual GPU hardware.
- GPU partitioning: The physical GPU is segmented into several dedicated slices.
- Virtual machines: Each VM is assigned its own vGPU.
- Dedicated VRAM: Every vGPU contains its own allocated VRAM.
- Isolation: Users operate strictly within their allocated GPU resources, preventing access to other users' vGPUs.
The specific number and size of vGPUs available are determined by the physical GPU model and the virtualization technology in use.
vGPU vs a Dedicated GPU
| Features | Dedicated GPU | vGPU |
|---|---|---|
| GPU allocation | One user or VM utilizes the physical GPU. | Multiple users or VMs share one physical GPU via separate vGPUs. |
| VRAM | The user accesses the GPU's available VRAM. | Each vGPU is allocated its own specific VRAM. |
| Users per GPU | Typically one. | Multiple, contingent on the GPU and configuration. |
| Best suited for | Workloads requiring substantial GPU resources. | Multiple workloads needing dedicated portions of a GPU. |
A dedicated GPU is more appropriate when a workload requires the majority or entirety of the card's resources. Conversely, vGPU technology is ideal when multiple users need GPU acceleration but do not require access to a full physical GPU individually.
What Can You Use a vGPU For?
vGPUs support a wide array of workloads that benefit from GPU acceleration. The suitable vGPU size is dictated by the specific software and workload requirements.
- AI and machine learning workloads
- 3D applications and engineering software
- Video editing
- Software development leveraging GPU acceleration
- Remote workstations
- Cybersecurity and other technical workloads
For demanding tasks such as large AI models, complex video projects, or intensive 3D applications, the amount of available VRAM is a critical consideration when selecting a GPU or vGPU configuration.
Why Use vGPUs in Cloud Desktops?
Cloud desktop environments can utilize vGPUs to deliver GPU-accelerated virtual machines to multiple users from the same physical hardware. This maximises GPU utilisation, particularly when individual users do not require the full capacity of a single card.
For example, a team can utilise separate virtual desktops while sharing the resources of a physical GPU through dedicated vGPU allocations. This ensures each user retains their own virtual GPU and isolated VRAM, rather than operating within a single shared desktop environment.
Try on DaDesktop
DaDesktop offers cloud desktops with dedicated GPUs and vGPU options, catering to workloads that demand GPU acceleration. Learn more about DaDesktop cloud GPU desktops.