Supermicro is transport its new GPU techniques — a groundbreaking portfolio primarily based on the Nvidia reference structure. That includes the superior Nvidia GH200 Grace Hopper and CPU superchip, the brand new modular structure permits enhanced AI infrastructure and improved computing.
The brand new design not solely affords scalability however flexibility for present and future GPUs, DPUs, and CPUs. With advances in liquid-cooling expertise, the servers supply high-density configurations and speedy processing. Supermicro assures supply of hundreds of rack-scale AI servers each month, with plug-and-play compatibility, marking a big step ahead for the trade.
Charles Liang, President and CEO of Supermicro, stated, ‘Supermicro’s NVIDIA MGX-based options present that our building-block technique permits us to carry the newest techniques to market shortly and are probably the most workload-optimised within the trade.’ He additional acknowledged, ‘Working with NVIDIA is crucial to speed up time to marketplace for enterprises growing new AI-enabled purposes, simplifying deployment, and lowering environmental affect.’
Ian Buck, NVIDIA’s vice chairman of hyperscale and HPC, recognised the significance of the collaboration between NVIDIA and Supermicro in delivering high-performing AI techniques.
Supermicro’s new product line addresses the distinctive thermal, energy, and mechanical challenges of AI-based servers. The in depth listing of choices contains ARS-111GL-NHR, ARS-111GL-NHR-LCC, ARS-111GL-DHNR-LCC, ARS-121L-DNR, ARS-221GL-NR, and SYS-221GE-NR amongst others, every designed to cater to totally different wants.
Technical specs reveal the spectacular capabilities of the brand new GPU techniques. The Supermicro’s 1U NVIDIA MGX techniques, for instance, have as much as 2 NVIDIA GH200 Grace Hopper Superchips. The memory-coherent, high-bandwidth, low-latency NVIDIA-C2C interconnects the CPU, GPU, and reminiscence at 900GB/s, seven occasions sooner than PCIe 5.0. Supermicro assists companies in growing effectivity and managing their knowledge apps successfully.
Alongside the high-performance machines, Supermicro additionally affords NVIDIA networking to safe and speed up AI workloads on the MGX platform. This features a mixture of NVIDIA BlueField-3 DPUs and ConnectX-7 adapters.
