NVIDIA Data Center GPUs for Servers

Accelerate your most demanding HPC and hyperscale data center workloads with NVIDIA® Data Center GPUs. Data scientists and researchers can now parse petabytes of data orders of magnitude faster than they could using traditional CPUs, in applications ranging from energy exploration to deep learning. NVIDIA’s accelerators also deliver the horsepower needed to run bigger simulations faster than ever before. Plus, NVIDIA GPUs deliver the highest performance and user density for virtual desktops, applications, and workstations.

Features HGX B300 RTX PRO 6000 (Server) H200 NVL L40S
FP32 75 TFLOPS 120 TFLOPS 60 TFLOPS 91.6 TFLOPS
GPU Memory 270 GB HBM3E 96 GB GDDR7 with ECC 141GB HBM3e 48GB GDDR6 with ECC
GPU Memory Bandwidth 7.7 TB/s 1597 GB/s 4.8TB/s 864GB/s
Power Consumption Configurable up to 1,100 W Up to 600W (Configurable) Up to 600W (Configurable) 350W
Multi-Instance GPUs 7 Up to 4 MIGs @ 24GB Up to 7 MIGs @16.5GB each N/A
Form Factor SXM PCIe Dual Slot PCIe Dual Slot PCIe Dual Slot

CONTACT US TO PURCHASE YOUR Customized NVIDIA DATACENTER SOLUTIONs TODAY

Resources