Wiwynn made a significant impact at NVIDIA GTC 2024 by unveiling its cutting-edge rack-level AI computing solutions featuring NVIDIA GB200 NVL72, alongside its liquid-cooled management system, UMS100.
The showcased NVIDIA GB200 Grace™ Blackwell Superchip supports the latest NVIDIA Quantum-X800 InfiniBand and Spectrum™-X800 Ethernet platforms. This innovation catapults generative AI to trillion-parameter scale, delivering 45x faster real-time large language model (LLM) inference, 40X lower TCO, and 40X less energy consumption. The GB200 Superchip can scale up to GB200 NVL72, a 72-GPU NVIDIA® NVLink®-connected GPU in a liquid-cooled rack. It acts as a single massive GPU with 30 TB of fast memory, providing unprecedented computing density for the next generation of large-scale generative AI.
As an NVIDIA Time To Market (TTM) partner, Wiwynn is among the first in line with NVIDIA GB200 NVL72 readiness, which aims to meet customer demands for enhanced performance, scalability, and diversity in data center infrastructures. Additionally, Wiwynn's SV7200H servers, equipped with GH200 superchips, underline the company's engineering prowess in high-speed data transmission and power efficiency, further complemented by its purpose-built liquid cooling management system, UMS100.
This comprehensive suite of offerings reflects Wiwynn's dedication to driving innovation in AI computing while ensuring adaptability and sustainability in data center solutions for the Gen AI era.