#Huawei #United Arab Emirates #Dubai #AI Data Center #Digital Era

Huawei Introduces AI Data Center Solution

On May 13, 2025, the Global Data Center Facility Summit 2025 convened in Dubai with a focus on "Powering the Digital Era Forward." The event gathered over 500 leaders, technical experts, and partners from the intelligent computing industry, providing a platform to discuss new growth opportunities and developmental pathways in the AI-driven data center sector. During the summit, Huawei unveiled its AI Data Center (AI DC) solution, a significant step toward ushering in a new era of intelligent computing.

Charles Yang, the Senior Vice President of Huawei and President of Global Marketing, Sales, and Services at Huawei Digital Power, opened the summit by emphasizing the golden age of computing infrastructure. He projected that global AI DC capacity would exceed 100 GW by 2028, predicting a market space of over $600 billion in the energy infrastructure sector alone.

Yang highlighted three major challenges that currently hinder AI DC developments: the need for higher reliability, faster deployment, and increased energy demands. According to Huawei, the future AI DC landscape will follow three significant trends:

1. Architectural Innovation for Improved Reliability: Enhancing reliability remains the industry’s top priority. Huawei believes that isolated architecture in power supply systems can vastly improve system dependability.

2. Accelerating AI DC Construction through Modular and Prefabricated Architecture: Huawei’s Outdoor PowerPOD implements a fully decoupled architecture to achieve one power system per container, significantly reducing time-to-market (TTM).

3. Enhancing Energy Efficiency through Green Power Distribution and Optimization: Utilizing green energy sources, through direct green power sources and campus PV development, while AI-based optimizations can increase energy efficiency in DCs.

In a session titled "Building a Solid Foundation for the Intelligent Computing Era," Bob He, Vice President of Huawei Digital Power, addressed various challenges faced in constructing smart computing centers. Despite the rising construction of these centers, they encounter issues related to rapid delivery, flexible deployment, and low total cost of ownership (TCO). He proposed the RASTM framework as foundational principles for AI DC construction:

• - Reliable: Ensuring end-to-end reliability throughout the entire lifecycle of AI DCs, spanning products, architecture, intelligent management, and professional services.
• - Agile: Flexibility is crucial for AI DCs to effectively deal with uncertainties in AI service deliveries. The Outdoor PowerPOD supports on-demand distribution and phased capacity increase under a completely decoupled architecture.
• - Sustainable: Improving power supply efficiency is vital. Emphasizing parallel system efficiency over module efficiency ensures operational optimization, with Huawei’s modular UPS5000-H achieving 97.5% efficiency for a single unit and up to 99.1% in eco mode for parallel systems.

Through dependable products, comprehensive lifecycle services, and expert consulting and design capabilities, Huawei aims to aid its partners in developing smart computing centers characterized by quick delivery, widespread reliability, optimized TCO, and robust flexibility, thereby propelling the digital era forward.

During the summit, Huawei, in collaboration with the International Telecommunication Union (ITU) and supported by Alibaba Cloud, China Mobile Group Design Institute, China Telecom, and China Unicom, published a white paper summarizing seven key aspects for enhancing the safety of lithium-ion battery applications in data centers. This initiative is aimed at boosting the industry's path towards high reliability and quality development.

Huawei Digital Power remains committed to innovation, striving to push technological boundaries and lead the deep exploration of the DC industry with pioneering products and solutions that contribute to the intelligent world and advance the digital age.