SK Hynix and NVIDIA are collaborating on the development of next-generation memory to build an essential infrastructure for the artificial intelligence (AI) era, known as the "AI factory." The two companies aim to establish a long-term technology partnership that integrates AI throughout the semiconductor design and manufacturing processes, accelerating their efforts to lead the AI semiconductor ecosystem.
On June 8, SK Hynix announced an expansion of its strategic cooperation with NVIDIA to enhance global AI infrastructure. Building on years of collaboration, the companies will jointly develop next-generation memory optimized for AI factories and work together on innovations in semiconductor design and manufacturing.
The focus of this partnership is the development of memory for next-generation AI systems. SK Hynix plans to co-develop memory for NVIDIA's upcoming AI supercomputer, "Vera Rubin," as well as for the Vera CPU, RTX Spark PC, and Jetson Thor robotics platform. The strategy also includes expanding their collaboration into personal AI and physical AI markets.
Considering the long development cycles of advanced memory, the companies will establish a stable supply chain. They plan to timely supply next-generation memory in line with NVIDIA's AI computing platform roadmap to meet the growing demand for global AI data center investments.
The collaboration will also extend into semiconductor development. SK Hynix is enhancing semiconductor simulation performance using NVIDIA's CUDA-X library and PhysicsNeMo framework. The goal is to improve the speed and accuracy of key development processes, including technology computer-aided design (TCAD) for analyzing process characteristics and computational lithography for implementing ultra-fine circuits.
The two companies plan to further expand their collaboration into electronic design automation (EDA) and overall simulation. This initiative aims to create a "triangular cooperation system" involving semiconductor manufacturers, AI platform companies, and EDA software firms to boost semiconductor development productivity.
In the field of manufacturing innovation, advancements in digital twin technology are being pursued. SK Hynix is utilizing NVIDIA's Omniverse and OpenUSD technologies to create a three-dimensional virtual representation of actual semiconductor factories, allowing for real-time analysis and optimization of production environments. They also plan to establish an autonomous manufacturing system linked to AI-based work processes in the future.
Jensen Huang, founder and CEO of NVIDIA, stated, "The AI factory is the engine of the next industrial revolution, and advanced memory is key to its performance. SK Hynix has been an outstanding partner, playing a crucial role in providing advanced memory technology for NVIDIA's AI computing platform. Together, we will co-develop next-generation memory for AI factories and support the acceleration of global AI infrastructure expansion from frontier model training to agentic AI and physical AI."
Chey Tae-won, Chairman of SK Group, remarked, "This partnership demonstrates the depth of collaboration that SK Hynix and NVIDIA have built over the years. By jointly developing next-generation memory for AI factories and applying AI to semiconductor design and manufacturing, we will shape the future of AI infrastructure together."
* This article has been translated by AI.
Copyright ⓒ Aju Press All rights reserved.