Compute

Key Features Two hot-pluggable systems (nodes) in a 1-OU form factor. Each node supports the following:

Key Features Four hot-pluggable systems (nodes) in a 2 OU form factor. Each node supports the following:

5OU, 16-GPU, Ultra High-Density Liquid Cooled System AMD Instinct™ MI325X GPU, Advanced Direct Liquid Cooled GPU Solution for AI, Inference, and Training ⦁  2x CPU Node: AMD EPYC™ 9005 Processors Dual Socket ⦁  2x GPU Tray:…

2OU2N AMD SP5 High Density Multi-Node Server with All-Flash Storage for up to 480TB Per Node Per Node ⦁  Dual socket AMD EPYC™ 9005 series processors, TDP up to 300W ⦁  24 x DDR5 DIMM slots,…

Arm’s AGI CPU reference server is a high-density, scale-out platform designed for cloud and AI infrastructure, featuring a dual-CPU blade architecture that maximizes core count and efficiency per rack. Built to support both single- and dual-socket…

Key Applications Key Features

Rapid.Space BBUs are the “all-in-one” (DU/CU) building blocks to deploy SimpleRAN 4G/5G public or private radio access networks. Available models rely on Intel Xeon and AMD Epyc CPUs. These support 19 inch telecom rack (ESA) and cloud native…

NVIDIA GB200 NVL72, 36 Grace™ CPUs and 72 Blackwell GPUs, NVIDIA NVLink™ Switch Tray, for Real-Time Trillion-Parameter Inference and Training ⦁  Per Rack: – 18 x GB200 compute trays + 9 x NVLink™ switch trays –…

8U 2-Socket NVIDIA HGX™ B300 8-GPU System Built for the Age of AI Reasoning for Every Data Center ⦁  NVIDIA HGX™ B300 8-GPU with NVIDIA NVSwitch™, 2.3TB HBM3e (8 x 288GB), TDP up to 1,200W ⦁…

8U 2-Socket NVIDIA HGX™ B200 8-GPU System Propelling The Data Center into A New Era of Accelerated Computing and Generative AI ⦁  NVIDIA HGX™ B200 8-GPU with NVIDIA NVSwitch™, 1.4TB HBM3e (8 x 180GB), TDP up…