Nvidia is making waves in the tech industry with the announcement of their new Blackwell architecture-powered systems. These systems, featuring Grace CPUs, Nvidia networking, and infrastructure, are designed for enterprises to build AI factories and data centers. The highlight of this new technology is the Nvidia GB200 Grace Blackwell Superchip, which offers exceptional performance gains and a 30 times increase in performance for LLM inference workloads compared to previous iterations.
The Blackwell platform is set to advance the next wave of generative AI, and Nvidia has partnered with top computer manufacturers including ASRock Rack, Asus, Gigabyte, Ingrasys, Inventec, Pegatron, QCT, Supermicro, Wistron, and Wiwynn to deliver cloud, on-premises, embedded, and edge AI systems using Nvidia GPUs and networking.
Nvidia CEO Jensen Huang expressed his excitement about the partnership, stating that “the next industrial revolution has begun” and that companies are partnering with Nvidia to shift traditional data centers to accelerated computing and build AI factories. This collaboration aims to produce a new commodity: artificial intelligence.
To address various application needs, the offerings will range from single to multi-GPUs, x86 to Grace-based processors, and air to liquid-cooling technology. Additionally, the Nvidia MGX modular reference design platform now supports Blackwell products, enabling the development of systems of different sizes and configurations.
Asus chairman Jonney Shih highlighted their collaboration with Nvidia to take enterprise AI to new heights with their powerful server lineup. By using Nvidia’s MGX and Blackwell platforms, Asus can craft tailored data center solutions for various customer workloads.
The GB200 NVL2, part of the Blackwell product lineup, is ideally suited for emerging market opportunities such as data analytics. It offers high-bandwidth memory performance and significantly improves data processing speed while providing better energy efficiency compared to x86 CPUs.
Nvidia’s MGX architecture is supported by AMD and Intel, who plan to deliver their own CPU host processor module designs. This partnership allows server system builders to save development time while ensuring consistency in design and performance.
Nvidia’s partner ecosystem includes TSMC, a leading semiconductor manufacturer, as well as global electronics makers. These partnerships enable the development and deployment of new data center infrastructure, further accelerated by Blackwell technology, Nvidia Quantum-2 or Quantum-X800 InfiniBand networking, Nvidia Spectrum-X Ethernet networking, and Nvidia BlueField-3 DPUs.
Taiwan is also embracing the Blackwell platform, with leading companies like Chang Gung Memorial Hospital and Foxconn adopting it for their own businesses. Chang Gung Memorial Hospital plans to use Blackwell to advance biomedical research and enhance patient care, while Foxconn aims to develop smart solution platforms for AI-powered electric vehicles and robotics.
The chairman of Quanta Computer, Barry Lam, expressed his excitement about joining Nvidia on this journey to define the next era of generative AI. Supermicro CEO Charles Liang highlighted their ability to quickly deliver game-changing Nvidia AI platform-based products to AI factories worldwide.
The partnership between Nvidia and these companies will shape and define a new chapter of AI, revolutionizing various industries and pushing the limits of semiconductor innovation. With their groundbreaking GPUs and the Blackwell architecture, Nvidia is at the forefront of AI technology.