Data Center Cooling Innovation: Water Resource Management for Competitiveness and Sustainability in the AI Era

~13 min read

Data Center Cooling Innovation: Water Resource Management for Competitiveness and Sustainability in the AI Era

The Importance of Data Center Cooling Technology in the AI Era

In recent years, Artificial Intelligence (AI) technology has undergone remarkable advancements, driving innovation across all sectors of society. This progress in AI naturally underscores the increasing importance of data centers, which store, process, and analyze data. With the emergence of hyperscale AI models like ChatGPT, data centers have evolved beyond mere data storage facilities to become core infrastructure for AI computation.

AI models learn from vast amounts of data and perform complex computations, leading to very high heat generation from servers within data centers. This intense heat can result in reduced server performance, system instability, and even equipment failure. Therefore, establishing a stable and effective cooling system is essential for efficient data center operation. Data center cooling technology plays a crucial role not only in lowering server temperatures but also in enhancing the overall energy efficiency of the data center and reducing operational costs.

South Korea boasts a world-class IT infrastructure and is actively engaged in AI technology development. Consequently, the domestic data center market is experiencing rapid growth, with increasing interest and investment in data center cooling technologies. The South Korean government, through its Digital New Deal policy, actively supports data center construction and operation, and is implementing various policies to enhance data center energy efficiency. For example, the Ministry of Trade, Industry and Energy is developing measures to improve data center energy efficiency and promoting support programs for the adoption of high-efficiency cooling technologies. Additionally, the Ministry of Environment provides guidelines to reduce data center water consumption and encourages the implementation of eco-friendly cooling technologies.

Data center cooling technology is a critical infrastructure supporting the advancement of various fields such as AI, cloud computing, and big data. Therefore, data center operators must maintain data center performance and ensure stable operations by adopting innovative cooling technologies. Furthermore, they must fulfill their social responsibility by building sustainable cooling systems that enhance energy efficiency and minimize environmental impact.

South Korea’s data center market is expected to continue its sustained growth, and data center cooling technology will play an even more crucial role. Data center operators must strengthen their competitiveness through continuous technological innovation and investment, leading the digital transformation in the AI era.

As of 2023, South Korea’s data center market size is estimated at approximately 4 trillion Korean Won (KRW), showing an average annual growth rate of over 15%. This growth is attributed to various factors, including the advancement of AI technology, the expansion of cloud services, and the acceleration of digital transformation among businesses. Particularly, with the activation of the non-face-to-face economy following the COVID-19 pandemic, data traffic surged, leading to increased demand for data centers. Additionally, the government’s Digital New Deal policy is promoting increased investment in data center construction and operation.

Water Resource Management: A New Criterion for Data Center Site Selection

In the past, the most critical criterion for data center site selection was a stable power supply. Data centers consume vast amounts of electricity, and any disruption in power supply can lead to operational issues. Therefore, data center operators preferred regions with well-established power grids, and areas with lower electricity costs were also considered attractive locations.

However, more recently, ‘water resource management’ has emerged as a critical limiting factor in data center site selection. Data center cooling systems consume enormous quantities of water, which translates into an environmental burden. Particularly, as South Korea is classified as a water-stressed country, data center water consumption is recognized as a serious issue. Therefore, data center operators must enhance their water management capabilities and explore ways to minimize water usage.

While power supply was a primary criterion for data center site selection in the past, the availability of stable water resources is now an even more crucial factor. Data center operators must thoroughly review water resource availability when selecting a data center site and actively consider adopting technologies to reduce water consumption. For example, data center operators can implement various cooling technologies such as air cooling, immersion cooling, and waste heat recovery systems to minimize water usage.

According to recently published data, water resource management efficiency is being utilized as a key indicator in evaluating data center sustainability. Data center operators must strengthen their ESG (Environmental, Social, and Governance) management and strive to improve water resource management efficiency to fulfill their social responsibilities. Furthermore, governments should promote policies to reduce data center water consumption and encourage the adoption of eco-friendly cooling technologies.

As of 2023, South Korea is classified as one of the OECD countries with a high water stress index, making data center water consumption an even more critical concern. Particularly during summer, water usage by data centers surges due to cooling system operations, which can exacerbate water scarcity in local communities. Therefore, data center operators must recognize the importance of water resource management and actively pursue efforts to reduce water consumption.

For example, Naver’s ‘Gak’ data center, located in Chuncheon, Gangwon Province, has implemented a rainwater harvesting system to self-supply the water needed for data center cooling. Similarly, Kakao’s data center in Seongnam, Gyeonggi Province, utilizes a wastewater recycling system to purify and reuse wastewater generated by the data center, thereby reducing water consumption. These examples demonstrate data center operators’ proactive efforts in water resource management.

Water Conservation Through Cooling Technology Innovation

Various technologies are being developed to reduce data center water consumption. These technologies can be broadly categorized into air cooling, immersion cooling, and waste heat recovery systems.

Air Cooling: Air cooling utilizes external air to cool servers. This method is effective for water conservation as it does not consume water. However, air cooling has drawbacks such as lower cooling efficiency compared to immersion cooling and potential noise generation. Recently, various technologies have been developed to enhance the efficiency of air cooling. For instance, technologies for filtering external air to remove dust and improve cooling efficiency are being developed. Additionally, techniques to optimize server placement to enhance airflow are also under development.

Immersion Cooling: Immersion cooling involves directly submerging servers in a dielectric fluid to maximize cooling efficiency. This method is effective for water conservation as it does not use water, and it offers higher cooling efficiency compared to air cooling. However, immersion cooling has disadvantages such as high initial investment costs and the need for coolant management. Recently, various technologies have been developed to mitigate the drawbacks of immersion cooling. For example, technologies to prevent coolant evaporation and enhance coolant circulation efficiency are being developed. Furthermore, eco-friendly coolants are being developed to minimize environmental impact.

Waste Heat Recovery Systems: Waste heat recovery systems are technologies that capture waste heat generated by data centers and repurpose it as heating energy. This method contributes to increasing data center energy efficiency while minimizing environmental impact. Waste heat recovery systems can be implemented in various forms depending on the data center’s location, scale, and cooling method. For example, waste heat generated by data centers can be used for district heating or for cultivating greenhouse crops.

These technologies contribute to both enhancing data center energy efficiency and minimizing environmental impact. Data center operators must adopt cooling technologies tailored to their data center’s specific characteristics to reduce water consumption and achieve sustainable data center operations.

In South Korea, KT has adopted immersion cooling technology to enhance its data center’s energy efficiency. Through immersion cooling, KT has reduced power consumption by over 30% compared to traditional air-cooled methods, contributing to lower carbon emissions from its data centers. Similarly, LG Uplus has implemented a waste heat recovery system, utilizing waste heat generated by its data centers for district heating. LG Uplus’s waste heat recovery system helps reduce district heating energy costs and enhances the energy self-sufficiency of local communities.

Gradient’s Hyperscale Data Center Water Treatment Contract

Gradient recently secured a contract for the design and construction of a water treatment facility for a hyperscale data center located in Didcot, Oxfordshire, UK. This is a prime example illustrating the critical importance of water resource management in data centers. Gradient plans to significantly reduce the data center’s water consumption and environmental impact by integrally providing water treatment, recycling, and Zero Liquid Discharge (ZLD) technologies.

Gradient’s ‘Site-Wide Peace of Mind’ approach is expected to play a crucial role in providing stable water resource management solutions to data center operators. Gradient plans to apply various technologies to reduce the data center’s water consumption. For example, Gradient intends to use Reverse Osmosis (RO) technology to purify wastewater generated by the data center and reuse the purified water in the cooling system. Additionally, Gradient plans to use Evaporation Concentration technology to concentrate wastewater and treat the concentrated wastewater as solid waste. Through these technologies, Gradient will minimize data center wastewater discharge and contribute to reducing water consumption.

According to industry analysis, hyperscale data centers consume significantly more water than conventional data centers. Therefore, hyperscale data center operators must dedicate greater efforts to water resource management. Gradient’s water treatment solutions are expected to provide stable and efficient water resource management solutions for hyperscale data center operators.

Water treatment facilities are a critical component for the stable operation of data centers. These facilities supply the water required for data center cooling systems and treat wastewater generated by the data centers. Consequently, data center operators are investing heavily in the design, construction, and operation of water treatment facilities.

In South Korea, Samsung Engineering has secured a competitive edge in the data center water treatment facility market. Based on its experience in the design, construction, and operation of data center water treatment facilities, Samsung Engineering is expanding its orders in both domestic and international data center markets. For example, Samsung Engineering has built SK Hynix’s data center water treatment facility and supports its stable operation. Furthermore, Samsung Engineering is pursuing water treatment facility contracts in overseas data center markets.

Customized Water Treatment Solutions for Data Centers

Gradient provides customized water treatment solutions that consider the specific characteristics of data centers. Data center cooling systems demand high stability, and water quality management is also extremely important. Gradient offers technologies that meet the water quality standards required for data center cooling equipment operation while minimizing water usage.

Gradient applies various water treatment technologies to supply the water needed for data center cooling systems. For example, Gradient uses Reverse Osmosis (RO) technology, Ion Exchange technology, and Ultraviolet (UV) disinfection technology to purify the water required for data center cooling systems. Additionally, Gradient employs various technologies to recycle wastewater generated by data centers. For instance, Gradient uses Membrane Separation technology, Biological Treatment technology, and Chemical Treatment technology to purify wastewater from data centers and reuse the treated water in the cooling system.

By purifying and reusing wastewater generated by data centers through wastewater recycling systems, water consumption can be further reduced. These customized solutions significantly contribute to enhancing data center sustainability.

Data center operators must adopt customized water treatment solutions tailored to their data center’s specific characteristics to reduce water consumption and enhance data center sustainability. Furthermore, data center operators should strive to improve the operational efficiency of water treatment facilities. For example, by analyzing operational data from water treatment facilities, they can improve operational efficiency and reduce maintenance costs.

In South Korea, Coway has secured a competitive edge in the data center water treatment solution market. Coway provides solutions for supplying water required by data center cooling systems and treating wastewater generated by data centers. Coway applies various water treatment technologies to meet data center water quality standards and offers stable water treatment solutions to data center operators. For example, Coway provides water treatment solutions to Naver’s ‘Gak’ data center, supporting its stable operation.

Sustainable Growth Strategies for Data Centers in the AI Era

The advancement of AI technology is leading to increased demand for data centers, which in turn can exacerbate energy consumption and environmental issues. Therefore, data centers must establish sustainable growth strategies. They need to minimize their environmental impact through the use of eco-friendly energy, the adoption of energy-efficient cooling technologies, and optimized water resource management.

Data center operators must strengthen their ESG management and strive to fulfill their social responsibilities. For example, data center operators should work to reduce their data centers’ energy consumption and carbon emissions. Additionally, they should endeavor to decrease water usage and minimize waste generation from their data centers.

Experts predict that data center sustainability will become a crucial factor determining corporate competitiveness. Data center operators must establish sustainable growth strategies and strengthen ESG management to enhance their corporate competitiveness.

The South Korean government is pursuing various policies to support the sustainable growth of data centers. For example, the government supports technology development to enhance data center energy efficiency and encourages the use of eco-friendly energy in data centers. Furthermore, the government provides guidelines to reduce data center water consumption and promotes policies to minimize waste generation from data centers.

Data center operators must actively participate in government policies and continuously pursue efforts to enhance data center sustainability. Moreover, data center operators should share information and collaborate to improve data center sustainability.

The role of data centers will become even more critical in achieving the 2050 carbon neutrality goal. Data center operators must actively strive to increase data center energy efficiency and utilize eco-friendly energy to meet carbon neutrality targets. Additionally, data center operators should invest in technology development to reduce data center carbon emissions and promote policies aimed at lowering these emissions.

The Future of Data Centers: Eco-Friendly, High-Efficiency, and Sustainable

The future of data centers can be summarized by three key themes: eco-friendliness, high efficiency, and sustainability. Data centers must reduce energy consumption, minimize environmental impact, and operate in a sustainable manner. To achieve this, data center operators must adopt innovative technologies and embrace eco-friendly operational practices.

Data center operators must strive to enhance their data centers’ energy efficiency. For example, they can improve energy efficiency by implementing high-efficiency servers, advanced cooling systems, and energy management systems. Furthermore, data center operators should work to reduce their data centers’ energy consumption. For instance, they can decrease energy usage through server virtualization, storage virtualization, and network virtualization.

Data center operators must endeavor to minimize their data centers’ environmental impact. For example, they should strive to use eco-friendly energy and reduce carbon emissions from their data centers. Additionally, data center operators should work to decrease water usage and minimize waste generation from their data centers.

Governments and businesses must collaborate to expand policies and investments for the sustainable growth of data centers. Governments should support technology development to enhance data center energy efficiency and encourage the use of eco-friendly energy in data centers. Furthermore, governments should provide guidelines to reduce data center water consumption and promote policies to minimize waste generation from data centers.

Data center sustainability is a core challenge in the AI era, and efforts to address it will continue. Data center operators must strengthen their competitiveness through continuous technological innovation and investment, leading the digital transformation in the AI era.

Data centers have evolved beyond simple data storage facilities to become core infrastructure for AI computation. Therefore, data center sustainability is a crucial prerequisite for the advancement of AI technology. Data center operators must continuously pursue efforts to enhance data center sustainability and lead the digital transformation in the AI era.