NVIDIA and SK hynix have joined forces in a multi-year technology partnership aimed at the advancement of next-generation memory solutions that align with NVIDIA’s strategic AI infrastructure vision. This collaboration plays a crucial role in ensuring a steady supply of sophisticated memory, which is vital for navigating the lengthy development cycles and substantial capital investments necessary for the global expansion of AI factories.
The partnership will enable SK hynix to enter emerging markets that NVIDIA is cultivating, including AI infrastructure, personal AI, and physical AI. Together, they will co-develop memory technologies intended for NVIDIA's cutting-edge Vera Rubin AI supercomputers, Vera CPUs, RTX Spark-equipped PCs, and Jetson Thor robotic computing platforms.
In an innovative approach, both companies will leverage AI to enhance semiconductor chip design and manufacturing processes. They plan to utilize NVIDIA's CUDA-X libraries and the PhysicsNeMo framework to streamline semiconductor simulations, technological computer-aided design (TCAD) workflows, and internal engineering projects.
Moreover, SK hynix is focusing on the development of digital twins for its fabrication facilities by integrating NVIDIA Omniverse technology, along with OpenUSD scene optimization and the NVIDIA cuOpt decision optimization engine, to create fully autonomous operations within factories.
The announcement of this partnership marks a significant milestone in the long-standing collaboration between NVIDIA and SK hynix, which has already contributed to some of the globe's most sophisticated AI computing solutions. As Jensen Huang, NVIDIA’s founder and CEO, stated, “AI factories are the engines of the next industrial revolution, and advanced memory is essential to their performance.” He emphasized SK hynix's pivotal role in equipping NVIDIA's AI computing platforms with cutting-edge memory technologies.
Chey Tae-won, Chairman of SK Group, reflected on the years of collaboration that led to this agreement, highlighting its potential to revolutionize AI infrastructure through shared efforts in semiconductor design and production.
This strategic partnership not only addresses the demands of advanced memory supply amid the rapid scaling of AI factories but also fosters innovation in fields such as AI, ultimately shaping the future landscape of technology. By employing NVIDIA's powerful simulation tools, SK hynix is positioned to accelerate semiconductor design and manufacturing, setting the stage for collaborative advancements in chipmaking.
Additionally, the creation of factory digital twins enables visualization and optimization of complex semiconductor manufacturing processes. These digital twins support operational efficiency, enhancing the coordination of autonomous mobile robots and various other assets within the fabrication environment. The initiative also explores ways to integrate these digital twins with existing software and intelligent AI systems, improving data-driven decision-making and task automation within manufacturing operations.
