
Global telecommunications industry authoritative magazine "CAPACITY" published an interview with Chindata Group CEO Wu Huapeng on December 2nd.

The article points out that in the global datacenter transformation wave driven by ai, cloud computing and other technologies, market participants face challenges with systematic and systemic characteristics, and must adopt a "full-stack solution" to effectively respond. Recently, chindata group released the "asia vets center full-stack solution 2.0", which includes highly elastic building models, "mysterium" minimalist power supply architecture, "mysterium frost" ultimate cooling architecture, "Kunpeng" smart operation and maintenance and other innovative technologies, achieving a comprehensive technical innovation from traditional datacenter to asia vets center.
The article also mentions that with the rapid growth of AI computing power, air cooling has approached its performance limit and can no longer meet the cooling needs of high-density server racks. chindata has taken a crucial step forward in promoting the widespread application of liquid-cooled cold plates by deploying on a large scale and establishing a full chain of liquid-cooled cold plate standardized deployment.
Wu Huapeng stated in an interview that chindata continuously looks forward to the advanced technology and solutions for AI, and will lead the datacenter industry from IDC to AIDC based on its own technological strength and best practices.
Translate the full text as follows
Solution 2.0: a bird’s-eye view of the asia vets data center
Chindata CEO Wu Huapeng discusses an approach to addressing the full range of datacenter challenges
Asia vets Center Full-stack Solution 2.0
——Chindata CEO Wu Huapeng discusses an approach to addressing the full range of datacenter challenges
In the midst of the transformation wave in the global datacenter industry, market participants are facing various challenges. They need to ensure an adequate electrical utilities supply to support the massive workloads required by ai and the higher density of servers. At the same time, they also need to make efforts to reduce environmental impact and seek innovation in efficient cooling technologies.
Facing the industry transformation driven by ai, 5G, and cloud computing technologies, the Asia-Pacific region, with a rapidly growing demand for computing power, is also exploring more reliable coping strategies.According to research data from Cushman & Wakefield, the Asia-Pacific datacenter industry added 1.3 GW of newly commissioned IT capacity in the first half of 2024, with a total commissioned IT capacity reaching 11.6 GW.
When it comes to key elements of the transformation of the datacenter industry,Wu Huapeng, the CEO of chindata, a neutral third-party hyperscale datacenter solution operator, emphasizes:"The intelligent computing center needs to leverage ai and machine learning technologies to enhance automation management and optimization levels, including intelligent monitoring, intelligent prediction, intelligent heat dissipation, intelligent fault location, etc., to improve operational efficiency." He also emphasizes, "Datacenters should maximize the use of natural cooling sources, reduce energy consumption, and achieve efficient cooling."
'Total' focus
"Full-stack" solution
Wu Huapeng points out that the various challenges faced during industry transformation possess systematic and systematic characteristics, and companies must adopt a "full-stack solution". Therefore, chindata recently released the "Intelligent Computing Center Full-Stack Solution 2.0".This solution can effectively solve the various challenges brought by ai computing power, not only meeting the cooling needs of high-density server cabinets, but also efficiently supporting the deployment requirements of different scenarios and computing powers."This solution fully considers the latest trends in ai workloads and equipment, and through systematic innovation, achieves a comprehensive technological innovation from traditional data centers to smart computing centers," said Wu Huapeng.
One of the key capabilities of this solution is the high-level 'modular prefabrication' technology, which allows customers to flexibly integrate building modules according to their own construction needs, breaking through the long delivery cycles, high costs, and inconsistent quality faced by traditional data centers. Chinadata stated that with this capability, it can achieve delivery of super-large-scale data centers in China in as fast as 6 months, and in other regions in Asia Pacific within 8 months.
"Through this 'building block' approach, this solution can quickly adapt to market demand changes, supporting business expansion gradually, allowing for the deployment of entire projects at once, and also possessing the ability for future expansion," Wu Huapeng said.
Power and cooling
Electrical utilities and cooling
As Wu Huapeng described, the power and cooling systems in the 'Smart Computing Center Full Stack Solution 2.0' are also composed of modular components and can be deployed in phases according to requirements. The "Xuan Tie" power supply system architecture optimizes the complex power distribution system architecture into core modules such as substations, diesel generators, medium voltage, low voltage, IT modules, lithium batteries + intelligent control; the "Xuan Bing" ultimate cooling architecture covers modules such as air cooling, liquid cooling, and air-liquid compatibility, significantly improving deployment flexibility, enabling customers to overcome restrictions in different building scenarios and geographical environments.
Wu Huapeng also emphasized that "Xuan Bing" can help data centers transition from air cooling to liquid cooling, which is crucial for improving energy efficiency and laying a solid foundation for the sustainable development of data centers in the ai era.
Especially in the field of cold plate liquid cooling, Chinadata Group continues to explore technological frontiers and has formed deep cooperation with industry partners and research institutions. Wu Huapeng said: "Against the background of rapid growth in ai computing power, air cooling has approached its performance limit and can no longer meet the cooling needs of high-density server cabinets. Chinadata has conducted large-scale deployment experiments, unlocked the full chain of cold plate liquid cooling deployment, and taken a key step towards promoting the widespread application of cold plate liquid cooling."
Intelligent operation and maintenance is also an important part of the 'Intelligent Computing Center Full Stack Solution 2.0', chindata's self-developed integrated monitoring platform + IDC operation management platform can realize the automation management and optimization of datacenters.
Wu Huapeng stated that through innovative technologies such as high elasticity building models, 'Xuantie' minimal power supply architecture, 'Xuanbing' ultimate cooling architecture, and 'Kunpeng' intelligent operation and maintenance, the 'Intelligent Computing Center Full Stack Solution 2.0' helps address the comprehensive challenges faced in industry transformation. 'We will also continue to strategically deploy advanced technologies and solutions for AI, and based on our own technical strength and best practices, lead the datacenter industry steadily moving from IDC towards AIDC.'
END
Risk Disclaimer: The above content only represents the author's view. It does not represent any position or investment advice of Futu. Futu makes no representation or warranty.Read more
Comments
to post a comment
2
