Your Browser is Out of Date

Nytro.ai uses technology that works best in other browsers.
For a full experience use one of the browsers below
VDI Design Guide—Dual VDI and AI on Dell 15G vSAN Ready Nodes in a Healthcare Environment
Introduction Solution Architecture
Architecture overview
Solution architecture
Physical architecture overview
Nodes front and rear views NVIDIA L40 GPUs Physical network components Storage options Data center infrastructure Client components
Software
Overview VMware vSphere VMware Horizon Horizon clone technology vSAN software-defined storage NVIDIA vGPU AI components
Validation Sizing the Solution Summary References

Physical architecture overview

Home > Workload Solutions > Virtual Desktop Infrastructure > Design and Implementation Guides > VDI Design Guide—Dual VDI and AI on Dell 15G vSAN Ready Nodes in a Healthcare Environment > Physical architecture overview

Physical architecture overview

  • NVIDIA L40 GPUs

    The NVIDIA L40 GPU Accelerator is a full-height, full-length (FHFL), dual-slot, PCI Express Gen4 graphics solution based on the latest NVIDIA Ada Lovelace Architecture. The card is passively cooled and capable of 300 W maximum board power. Fourth-generation Tensor Cores deliver enhanced AI capabilities to accelerate visual computing workloads and groundbreaking performance for deep learning and inference applications. A PowerEdge R750 vSAN Ready Node can host a maximum of two NVIDIA L40 GPUs.

    Figure 6. NVIDIA L40 GPU
    Table 2. NVIDIA L40 GPU Specifications
    GPU Architecture Ada Lovelace
    CUDA Cores 18176
    RT Cores 142 - Gen 3
    Tensor Cores 568 - Gen 4
    Memory Size 48 GB GDDR6 with ECC
    Memory Peak Bandwidth Up to 864 GB/s
    FS32 Performance 16.3 TFLOPS
    Board Power 300 W
    Power Connector PCIe CEM5 16-Pin
    vGPU Software Support 15.1 or later