NVIDIA is the most popular technology company in recent times. The leading AI computing chips and explosive performance have driven the company’s market value to over a trillion dollars. Its highly respected CEO, Huang Renxun, also stated that “we are in the iPhone moment of AI”.
This means that AI applications represented by chatGPT have reached the turning point of takeoff. Nvidia believes that this is related to two factors: firstly, computing architecture, the new accelerated computing architecture driving a surge in computing power; The second is deep learning, which enables AI to easily transition between different modalities.
The trillion dollar data center market is undergoing profound changes driven by AI, with Nvidia’s GPUs and various high-performance chips providing strong support. At the same time, NVIDIA also plays an important role in the telecommunications industry, and its technology is the backbone driving the vigorous development of the telecommunications industry. The current telecommunications industry has entered the 5G era and is transitioning towards a network architecture centered around data centers. For example, in China, operators are the main force in the construction of the “East to West Computing” project, with a large number of data centers and millions of racks deployed, aiming for stronger computing capabilities.
Taking Japanese operator Softbank as an example, Nvidia recently announced that they are collaborating to create a groundbreaking platform based on the NVIDIA GH200 Grace Hopper superchip for generative AI and 5G/6G applications. SoftBank plans to adopt this platform in new distributed AI data centers across Japan.
Through cooperation with Nvidia, SoftBank will build a data center that can host generative AI and wireless applications on a multi tenant universal server platform, in order to reduce costs and become more energy-efficient. Compared to previous data centers, these new data centers will be more evenly distributed and able to handle both AI and 5G workloads simultaneously. This will enable them to better maintain peak operation while maintaining low latency and significantly reducing overall energy costs.
The demand for accelerated computing and generative AI is driving fundamental changes in data center architecture, “said Huang Renxun, founder and CEO of NVIDIA.” NVIDIA Grace Hopper is a revolutionary computing platform designed to handle and expand generative AI services. Like other visionary initiatives in the past, SoftBank is leading the world in creating a telecommunications network to host generative AI services.”

NVIDIA Grace Hopper and NVIDIA BlueField – 3 DPU will accelerate software defined 5G vRAN and generative AI applications without the need for customized hardware accelerators or dedicated 5G CPUs. In addition, NVIDIA Spectrum Ethernet switches equipped with BlueField-3 will provide high-precision time synchronization protocols for 5G.
As stated by Ren Haas, CEO of Arm, the future development of generative AI requires high-performance and energy-efficient computing. Nvidia has significantly increased the throughput of its 5G network and reduced energy consumption in massive computing environments through cooperation with SoftBank. It is reported that by introducing NVIDIA Aerial software into high-performance software defined cloud native 5G networks, 5G base stations will allow operators to dynamically allocate computing resources and achieve energy efficiency 2.5 times higher than competitors. This is of great significance for operators to develop digital services for industry customers.
Softbank is developing 5G applications for autonomous driving, AI factories, augmented reality and virtual reality, computer vision, and digital twins. For domestic operators, the above industry applications are also advancing – even faster. The adoption of stronger computing chips will have significant implications for the network transformation of operators.