Building Taiwan’s Largest Supercomputer with Nvidia Blackwell Chips
Introduction
Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More
Nvidia and Foxconn Collaboration
Nvidia and Foxconn are collaborating to build Taiwan’s largest supercomputer using Nvidia Blackwell chips. The project, known as the Hon Hai Kaohsiung Super Computing Center, was unveiled at Hon Hai Tech Day. It will be constructed around Nvidia’s Blackwell GPU architecture, featuring the GB200 NVL72 platform with 64 racks and 4,608 Tensor Core GPUs.
Supercomputer Performance
The supercomputer is expected to deliver over 90 exaflops of AI performance, making it the fastest in Taiwan. Foxconn plans to utilize this high-performance computing system to advance cancer research, develop large language models, and drive smart city innovations, positioning Taiwan as a global leader in AI-driven industries.
Foxconn’s Three-Platform Strategy
Foxconn’s “three-platform strategy” focuses on smart manufacturing, smart cities, and electric vehicles. The new supercomputer will support Foxconn’s initiatives in digital twinning, robotic automation, and smart urban infrastructure, bringing AI-assisted services to urban areas like Kaohsiung.
Construction has commenced on the supercomputer in Kaohsiung, with the first phase expected to be operational by mid-2025 and full deployment by 2026. The project integration with Nvidia technologies like Omniverse and Isaac robotics platforms will revolutionize manufacturing processes.
Nvidia is providing Blackwell AI chips to Foxconn for the new supercomputer.
Supercomputer Features
The supercomputer is powered by Nvidia’s Blackwell platform, offering exceptional AI computing power. The GB200 NVL72 platform combines advanced CPUs and GPUs to deliver high bandwidth and efficiency. Nvidia NVLink Switch enables the GPU system to function as a unified entity, ideal for training large AI models and real-time complex inference tasks.
Foxconn’s Advancements
Foxconn, the world’s largest electronics manufacturer, is leveraging Nvidia technologies to build digital twins of its factories and develop domain-specific large language models. Through initiatives in smart manufacturing, smart electric vehicles, and smart cities, Foxconn aims to revolutionize various industry sectors.
FAQs
What is the purpose of the supercomputer being built by Nvidia and Foxconn?
The supercomputer aims to advance AI research, support innovative projects in cancer research, large language models, and smart city developments in Taiwan.
How is Foxconn utilizing the supercomputer?
Foxconn plans to harness the supercomputer’s power for digital twinning, robotic automation, and smart urban infrastructure in areas like Kaohsiung.
When will the project be fully operational?
The first phase of the supercomputer is expected to be operational by mid-2025, with full deployment targeted for 2026.
Credit: venturebeat.com
