Recommended NVIDIA GPUs for NVIDIA vPC

Density-optimized GPUs are typically recommended for vPC deployments targeting knowledge workers running office productivity applications, video conferencing, or modern desktop environments such as Windows 10/11 or Linux distributions. These GPUs are ideal for maximizing the number of users per server and offer excellent cost-efficiency for large-scale VDI deployments.

The RTX PRO 6000 Blackwell Server Edition, a performance and density optimized GPU, is also recommended for vPC environments. It offers the flexibility and performance headroom needed to support more demanding users alongside typical vPC workloads, making it ideal for deployments that require both scalability and versatility.

Table 1 NVIDIA GPUs Recommended for vPC

	RTX PRO 6000 Blackwell Server Edition	NVIDIA A16
Number of GPUs / Boards [Architecture]	1 [Blackwell]	4 [Ampere]
RT Cores	188	40 [10 per GPU]
Memory Size	96 GB GDDR7	64 GB GDDR6 [16GB per GPU]
Form Factor	PCIe 5.0, Dual Slot FHFL	PCIe 4.0, Dual Slot FHFL
Power	600W	250W
Thermal	Passive	Passive
Optimized for	Performance and Density	Density and Cost per User
1B Users per Board with Homogeneous vGPU	Not supported on Blackwell	64 [16 Users per GPU]
2B Users per Board with Homogeneous vGPU	32 (48 with MIG-backed Time-Sliced vGPU)	32 [8 Users per GPU]
3B Users per Board with Homogeneous vGPU	32	20 [5 Users per GPU]

The NVIDIA RTX PRO 6000 Blackwell Server Edition supports mixed workloads and is well-suited for both vWS and vPC deployments. With 96 GB of GDDR7 memory, high AI and graphics throughput, and MIG capabilities, it allows for flexible allocation of GPU resources during business hours and compute-heavy tasks during off-hours. This makes it a powerful choice for organizations seeking performance and versatility in their virtualized environments.

The NVIDIA A16, based on the NVIDIA Ampere architecture, is a dual-slot FHFL card with 64 GB of total memory (4 GPUs with 16 GB each) and a 250 W passive design. It is optimized for high user density and delivers exceptional value in vPC environments targeting task and knowledge workers. The A16 enables IT teams to maximize data center efficiency by consolidating user sessions at scale. Its low power profile and density-focused design make it ideal for environments where cost-per-user and resource utilization are critical.

Note

For the ability to run mixed workloads with the A16 or RTX Pro 6000 Blackwell Server Edition, please note the appropriate software licenses within the NVIDIA Virtual GPU Software Packaging, Pricing, and Licensing Guide.

All current GPUs support ECC memory, and it’s enabled by default, which reduces the size of usable VRAM compared to what’s available when ECC is disabled. The physical GPU (i.e. not running vGPU) sees the same VRAM reduction when ECC is enabled. It is essential to resize your environment when switching to newer GPUs like the A16 or the RTX PRO 6000 Blackwell Server Edition.

Note

To maximize available VRAM on newer GPUs, ensure ECC memory is disabled. This step is crucial when transitioning from Maxwell, Pascal, and Turing architectures. However, this does not apply to Blackwell-generation GPUs, where ECC is built directly into the GDDR7 memory and is always enabled without reducing usable VRAM.