Guangdong Provincial Department of Industry and Information Technology Visits for Research

As artificial intelligence continues to drive unprecedented demand for high-density computing, liquid cooling has become a critical technology for next-generation data centers. On July 29, 2026, a delegation from the Guangdong Provincial Department of Industry and Information Technology, the Guangdong Electronic Information Industry Association, and the Shenzhen Municipal Bureau of Industry and Information Technology visited Lianli Liquid Cooling for an in-depth technical exchange and industry research.

The visit highlighted the growing importance of liquid cooling technologies in supporting AI infrastructure while strengthening collaboration between government agencies and technology enterprises to accelerate the development of China's intelligent computing ecosystem.

Showcasing Comprehensive Liquid Cooling Solutions

Led by General Manager Guo Wei and the company's management team, the delegation toured Lianli's product showroom and solution exhibition area, gaining a comprehensive understanding of the company's latest innovations in liquid cooling technology.

Lianli Liquid Cooling has established a complete product portfolio covering the entire AI data center cooling ecosystem, including:

• Containerized Liquid Cooling Data Centers
• Rack-Level Coolant Distribution Units (CDUs)
• Cabinet CDUs
• Liquid Cooling Server Racks
• CPU and GPU Cold Plates
• Cold Plate Liquid Cooling Systems
• Immersion Cooling Solutions

These products are designed to support the latest AI computing platforms, including NVIDIA GB300, GB200, B300, B200, H200, and H100 systems, as well as Intel Xeon 6 and AMD EPYC 9005 processors, providing reliable thermal management for AI clusters, HPC environments, and cloud computing facilities.

The delegation highly recognized the company's continuous investment in technology innovation and its expanding capabilities in AI infrastructure cooling.

Delivering High-Efficiency and Sustainable Cooling

With AI servers reaching rack power densities far beyond the limits of conventional air cooling, energy-efficient liquid cooling has become essential for modern data centers.

According to Lianli, its cold plate liquid cooling solutions can reduce data center Power Usage Effectiveness (PUE) to below 1.15, with optimized deployments achieving PUE values as low as 1.05. Compared with traditional air-cooled systems, overall energy consumption can be reduced by more than 30%.

The company also demonstrated its modular containerized liquid cooling solution. A single prefabricated 20-foot container is capable of supporting the cooling requirements of three full NVIDIA GB200 AI rack systems, significantly increasing computing density while shortening deployment time for AI infrastructure projects.

Lianli's products have obtained multiple international certifications, including CE, UL, RoHS, and CCS, and more than five million liquid cooling units have been deployed worldwide across over 100 countries and regions.

Productive Discussions on Industry Development

Following the facility tour, government representatives and Lianli executives held a roundtable discussion focusing on industry trends, technology development, supply chain collaboration, and policy support.

Lianli emphasized that cold plate liquid cooling remains its primary business and is expected to continue dominating the AI data center market over the coming years. Thanks to its mature technology, cost efficiency, and simplified maintenance, cold plate cooling has become the preferred solution for most AI server deployments.

The company also reported strong international business growth. Overseas markets now account for approximately 50%–60% of total sales, with major customers located in the United States, Canada, and Germany, while domestic projects continue to expand across Guangdong Province and northwestern China.

Looking ahead, Lianli plans to further increase cooling density within limited floor space. Its current technology can provide approximately 600 kW of cooling capacity within a 200-square-meter facility, with future development targeting megawatt-level cooling capacity in the same footprint.

Cold Plate Cooling Remains the Mainstream Choice

The discussion also covered immersion cooling, another emerging technology for AI infrastructure.

Lianli noted that while immersion cooling offers excellent thermal performance in certain scenarios, it still faces several practical challenges, including customized server design requirements, higher initial investment, complex maintenance procedures, and limited suitability for upgrading existing air-cooled data centers.

Additionally, some immersion cooling fluids remain relatively expensive and present environmental and lifecycle management considerations.

By comparison, cold plate liquid cooling offers a balanced combination of efficiency, reliability, lower deployment costs, and easier maintenance. As a result, it currently accounts for approximately 70% of the liquid cooling market, making it the preferred technology for today's AI data centers.

The company also highlighted the rapid progress of China's liquid cooling supply chain. With the exception of semiconductor chips, most key components—including pumps, heat exchangers, manifolds, and piping systems—can now be sourced domestically, strengthening the industry's competitiveness and resilience.

Government Support Accelerates Industry Growth

Government representatives expressed strong support for continued collaboration between industry associations, manufacturers, and technology providers.

They emphasized the importance of building stronger industrial partnerships across the Guangdong-Hong Kong-Macao Greater Bay Area, particularly in PCB manufacturing, AI server production, and liquid cooling equipment manufacturing.

By improving cooperation throughout the supply chain, the region aims to establish a globally competitive AI infrastructure ecosystem capable of supporting rapid industry expansion.

Favorable Policies Create New Opportunities

China's policy environment continues to provide strong momentum for the liquid cooling industry.

Recent national initiatives encourage the construction of energy-efficient AI data centers and promote advanced cooling technologies to reduce energy consumption while supporting the country's carbon neutrality objectives.

At the local level, Shenzhen's AI server industry development plan identifies cold plate liquid cooling, immersion cooling, and direct-to-chip cooling technologies as strategic priorities for the coming years.

Meanwhile, next-generation AI platforms continue to push rack power densities beyond the capabilities of traditional air cooling. As AI clusters move from approximately 10 kW per rack to 200–300 kW and beyond, liquid cooling is rapidly becoming the standard infrastructure for large-scale AI computing.

Industry analysts expect the global liquid cooling market to maintain strong growth over the next several years, driven by increasing investments in AI data centers, hyperscale cloud infrastructure, and high-performance computing.

Driving the Future of AI Liquid Cooling

Founded in 2014, Lianli Liquid Cooling is headquartered in Shenzhen, China, and specializes in advanced thermal management solutions for AI data centers, cloud computing facilities, and high-performance computing applications.

With continuous investment in research and development, an expanding international customer base, and a comprehensive portfolio of liquid cooling products, the company has established itself as one of China's leading providers of AI liquid cooling infrastructure.

The visit by Guangdong and Shenzhen industry officials reflects growing recognition of the strategic role that liquid cooling plays in enabling the next generation of AI computing.

As AI workloads continue to increase and sustainability becomes a core priority for data center operators worldwide, Lianli Liquid Cooling remains committed to delivering innovative, energy-efficient, and reliable liquid cooling solutions that support the future of intelligent computing.