NVIDIA and the world’s leading computer manufacturer have unveiled advanced systems based on the NVIDIA Blackwell architecture with Grace CPUs and NVIDIA networking to help enterprises build AI factories and data centers. According to the NVIDIA newsroom, this initiative aims to drive the next generation of generative AI innovation.
COMPUTEX Keynote Speech Highlights
During the COMPUTEX keynote, NVIDIA founder and CEO Jensen Huang announced collaborations with prominent industry players. Companies such as ASRock Rack, ASUS, GIGABYTE, Ingrasys, Inventec, Pegatron, QCT, Supermicro, Wistron, and Wiwynn will leverage NVIDIA GPUs and networking infrastructure to deliver cloud, on-premise, embedded, and edge AI systems.
Huang highlighted the beginning of the next industrial revolution, in which companies and countries are collaborating with NVIDIA to transform traditional data centers into accelerated computing hubs known as AI factories. This AI factory is set up for production. A.I On a large scale.
comprehensive service
We offer a variety of products ranging from single to multiple GPUs, x86 to Grace-based processors, and air-to-liquid cooling technologies to meet a variety of application requirements. The NVIDIA MGX™ modular reference design platform now supports Blackwell products, including the new NVIDIA GB200 NVL2 platform, designed for mainstream large language model inference, search augmentation generation, and data processing.
The GB200 NVL2 platform is particularly suited to emerging markets such as data analytics, where companies invest billions of dollars each year. Leveraging high-bandwidth memory performance through the NVLink®-C2C interconnect and a dedicated decompression engine, Blackwell architecture delivers up to 18x faster data processing and 8x more energy efficiency compared to x86 CPUs.
Modular Reference Architecture for Accelerated Computing
NVIDIA MGX provides a reference architecture that allows manufacturers to quickly and cost-effectively deploy more than 100 system design configurations. This flexibility allows manufacturers to choose GPUs, DPUs, and CPUs to handle a variety of workloads. Over 90 systems from over 25 partners are available or in development. This is a significant increase from last year’s 14 systems delivered by six partners. With MGX, you can reduce development costs by up to 3/4 and shorten development time to just 6 months.
AMD and Intel are also supporting the MGX architecture with plans to offer their own CPU host processor module designs. It includes the next-generation AMD Turin platform and Intel® Xeon® 6 processors with P-Cores (formerly codenamed Granite Rapids). These reference designs aim to save development time while ensuring design and performance consistency.
Ecosystem and partner contributions
NVIDIA’s comprehensive partner ecosystem includes TSMC, the world’s leading semiconductor manufacturer, and global electronics manufacturers that provide key components to AI factories. These components include server racks, power supplies, and cooling solutions from companies such as Amphenol, Asia Vital Components (AVC), Cooler Master, Colder Products Company (CPC), Danfoss, Delta Electronics, and LITEON.
This joint effort will enable us to rapidly develop and deploy new data center infrastructure to meet the needs of global enterprises. Blackwell technologies further accelerate these developments with NVIDIA Quantum-2, Quantum-X800 InfiniBand networking, NVIDIA Spectrum™-X Ethernet networking, and NVIDIA BlueField®-3 DPUs.
Enterprises can also access the NVIDIA AI Enterprise software platform, which includes NVIDIA NIM™ inference microservices, to create and run production-grade generative AI applications.
Adoption in Taiwan
In his keynote address, Huang highlighted that leading companies in Taiwan are rapidly adopting Blackwell to integrate AI into their businesses. Chang Gung Memorial Hospital plans to use the NVIDIA Blackwell computing platform to advance biomedical research and improve clinical workflows. Foxconn plans to develop smart solutions for AI-based electric vehicles and robot platforms as well as generative AI services.
Image source: Shutterstock
. . .
tag