AI and GPUs Redefine Edge Computing in 5G Networks | Hardware examples | List of hardware components with examples | Gpu comparison | Turtles AI

AI integration in edge computing is transforming mobile operators’ 5G networks, with the adoption of advanced GPUs to improve performance and energy efficiency.

Key Points:

• GPU Adoption in 5G Networks: Operators like NTT DOCOMO are deploying GPUs to accelerate 5G networks, improving efficiency and reducing costs.
• GPU-as-a-Service Services: Over 15 mobile operators are investing in GPU-equipped data centers to offer AI inference services to customers.
• Energy Efficiency and Cost Reduction: Using GPUs can reduce base station power consumption and operating costs by up to 50%.
• New Revenue Opportunities: The integration of AI and edge computing opens up innovative services in industries such as automotive, industrial, and consumer goods.

AI integration in edge computing is redefining mobile operators’ strategies for deploying 5G networks. In recent months, companies such as NVIDIA have introduced advanced GPU-based solutions, such as Grace Hopper and Grace Blackwell processors, to handle complex workloads in 5G radio access network (RAN) processing. At the same time, more than 15 mobile operators globally are investing in GPU-equipped data centers, offering “GPU-as-a-Service” services to their customers.

A notable example is NTT DOCOMO, which announced the launch of the first commercial GPU-accelerated 5G network. Working with Fujitsu and NVIDIA, the Japanese operator implemented a virtualized 5G Open RAN (vRAN) using the NVIDIA Aerial platform and NVIDIA BlueField converged accelerators. This solution has enabled total cost reductions of up to 30% and base station power consumption of up to 50%, compared to traditional implementations.

The adoption of GPUs in network infrastructure not only improves operational efficiency, but also opens up new revenue opportunities for operators. By offering real-time AI inference services, operators can meet the needs of industries such as automotive, manufacturing, and consumer goods, where AI and augmented reality applications require low latency and high processing capacity.

Deploying GPU infrastructure is a strategic choice for operators: while some consider deploying GPUs locally to handle specific workloads with low latency, others consider using regional or national data centers more efficient. These decisions depend on various factors, including capital costs (CAPEX), operating costs (OPEX), and energy savings associated with each approach.

Additionally, integrating AI into edge computing enables operators to offer innovative services, such as advanced video analytics, predictive maintenance, and real-time data processing, improving user experience and creating new revenue streams.

The adoption of GPU-based solutions in edge computing represents a significant step for mobile operators in the 5G era, offering benefits in terms of performance, energy efficiency and new business opportunities.