In response to the rapid global expansion of artificial intelligence infrastructure, LG has introduced a comprehensive suite of technologies designed to manage the immense power and cooling demands of next-generation data centers. The initiative, showcased at Data Centre World Asia 2025 in Singapore, integrates the capabilities of three of the company’s affiliates to offer a unified platform for building and operating more efficient, scalable, and sustainable AI facilities. This move addresses a market that is projected to grow by 24% between 2025 and 2034, reaching an estimated value of $221.4 billion.
The new portfolio, named the “One LG Solution,” combines advanced thermal management from LG Electronics, resilient power systems from LG Energy Solution, and expertise in data center design, construction, and operation from LG CNS. This collaborative approach aims to provide data center operators with a single, strategic partner for navigating the complexities of AI workloads, which generate significantly more heat and consume more power than traditional computing tasks. The integrated offerings range from innovative direct-to-chip liquid cooling to high-density battery storage and modular, containerized data center units.
A Unified Strategy for AI Infrastructure
The core of the new initiative is the integration of distinct strengths from three LG businesses into a single, cohesive offering. This strategy was developed to meet the growing global demand for AI infrastructure by positioning LG as a key partner for operators seeking to improve both efficiency and sustainability. The approach combines LG Electronics’ cutting-edge cooling technologies with LG Energy Solution’s battery systems and LG CNS’s extensive experience in data center architecture and management.
LG CNS, a prominent AI transformation company in South Korea, leads the design, construction, and operational management aspects of the joint venture. LG Electronics provides the critical thermal solutions required to handle the high-density heat loads produced by AI processors. To ensure operational continuity, LG Energy Solution contributes its advanced uninterruptible power supply (UPS) systems, which leverage its patented battery technology to deliver reliable energy. According to James Lee, President of the LG ES Company, the One LG Solution is “ready to reshape the global data center landscape” by supplying operators with everything needed to scale efficiently in the AI era.
Advanced Thermal Management Systems
A primary challenge in AI data centers is managing the intense heat generated by powerful processors running complex workloads. LG’s portfolio directly addresses this with a multi-pronged approach to thermal management that moves beyond traditional air cooling.
Direct-to-Chip and Liquid Cooling
Central to LG’s strategy is the use of direct-to-chip (DTC) cooling, a form of liquid cooling that is significantly more effective than air-based methods. This technology involves mounting cold plates directly onto heat-generating components like CPUs and GPUs. A specialized coolant circulates through channels in these plates, absorbing heat at the source before carrying it away to a Coolant Distribution Unit (CDU) to be chilled and recirculated. By directly transferring heat, DTC systems can be 50 to 1,000 times more efficient than air cooling, reducing the energy required for thermal management by 25–30%. This efficiency allows for higher server densities, enabling more computing power within the same physical footprint.
Innovations in Chillers and Air Handlers
Supporting these advanced liquid cooling systems are LG’s high-capacity, water-cooled chillers. These units feature oil-free magnetic bearing compressors, which improve reliability and further reduce energy consumption compared to conventional designs. The company’s offerings also include computer room air handlers (CRAHs) that manage airflow more effectively to enhance operational stability. For its CDUs, LG has developed high-efficiency inverter pumps and a modular design that allows for greater scalability and installation flexibility.
Exploring Immersion Cooling
Looking toward future demands, LG CNS is actively conducting proof-of-concept testing on immersion cooling, one of the most advanced thermal management techniques available. This method involves fully submerging server hardware in a thermally conductive but non-electrically conductive, or dielectric, fluid. The liquid directly absorbs heat from all components, eliminating the need for server fans and traditional heat sinks entirely. Immersion cooling offers superior thermal efficiency, enabling ultra-high-density deployments and potentially lowering a facility’s Power Usage Effectiveness (PUE) to less than 1.05, representing near-optimal energy efficiency.
Resilient and Efficient Power Solutions
The massive energy requirements of AI systems necessitate equally innovative power delivery and backup systems. LG Energy Solution provides the foundation for operational stability with a focus on efficiency and reliability.
Next-Generation UPS Technology
To guard against power fluctuations or outages, LG provides high-density uninterruptible power supply (UPS) systems. These systems use advanced battery technology to ensure a continuous and stable power supply, which is critical for maintaining uptime during demanding AI workloads. In response to the irregular power consumption patterns of AI data centers, the company is debuting a “Next Generation UPS Battery” capable of delivering more than double the power output of existing models. This high-output performance ensures reliable operation even under the most strenuous conditions.
Direct Current Grid Integration
To further boost efficiency, LG has developed an innovative direct current (DC) grid solution. In typical data centers, converting alternating current (AC) from the power grid to the DC used by servers results in significant energy loss, often around 25%, dissipated as heat. LG’s DC grid solution reduces this loss to approximately 10% by allowing systems like chillers to operate directly on DC power. This approach not only improves power efficiency but also contributes to lower overall operating costs and a reduced carbon footprint.
Intelligent Operations and Modular Design
Beyond cooling and power, LG CNS contributes expertise in data center architecture and intelligent operational management. This includes leveraging digital technologies to optimize performance and developing compact, scalable infrastructure.
The Compact ‘AI Box’ Concept
LG CNS has developed the “AI Box,” a compact, container-sized solution that integrates all core data center infrastructure components into a single, modular unit. This all-in-one approach is designed to streamline deployment and allow for rapid scalability. By consolidating cooling, power, and IT hardware into a prefabricated module, operators can build out or expand their AI capabilities more quickly and efficiently compared to traditional construction methods.
Digital Twin and Predictive Maintenance
To enhance operational control, the One LG Solution is augmented by a Data Centre Cooling Management System and Digital Twin technologies. A digital twin is a virtual model of a physical data center that allows operators to simulate changes, optimize performance, and predict potential issues before they occur. This enables predictive maintenance, where problems can be addressed proactively, minimizing downtime and improving the overall reliability of the facility.
Global Expansion and Market Response
LG is already applying its integrated solutions in strategic projects worldwide. The company is supplying the core systems for a hyperscale AI data center currently under construction in Jakarta, Indonesia. Furthermore, it has secured an agreement to provide its advanced cooling systems to a large-scale, net-zero AI data center being built in NEOM City, Saudi Arabia. These projects reflect a strong industry demand for the kind of energy-efficient and sustainable infrastructure LG aims to provide. With its unified approach, the company is positioning itself to be a key player in building the next generation of data centers required to power the ongoing AI revolution.
