LG Uplus plans to launch AI data center tours for customers and industry partners starting later this year
In sum – what you need to know:
LG Uplus launches cooling demo — The company opens a real-world demo room to test liquid cooling tech for high-density AI workloads at its major data center in Anyang, Korea.
Advanced cooling under review — Direct-to-chip and immersion cooling methods are being tested with help from partners like Vertiv, CoolIT Systems and LG Electronics.
Green data center strategy — LG Uplus plans to scale liquid cooling to its full facilities to reduce emissions, cut energy use and boost reliability in next-gen AI data centers.
Korean telco LG Uplus announced the launch of a next-generation liquid cooling demonstration room at its Pyeongchon 2 Internet Data Center (IDC) in Anyang, Gyeonggi Province. The move is part of its broader strategy to validate sustainable cooling technologies for AI data centers (AIDC).
The demo room showcases advanced cooling systems designed for high-heat, high-density AI server environments. The goal is to improve energy efficiency and reduce carbon emissions — key components of what the company refers to as a future-ready Green AI Data Center. By testing performance and stability in realistic settings, LG Uplus aims to reduce risks tied to GPU upgrades and large-scale infrastructure deployments.
The initiative brings together global cooling leaders including Vertiv and CoolIT Systems, as well as Korean firms such as LG Electronics and Global Standard Technology (GST). These companies are working jointly to validate commercialization potential using real operational data.
Two main technologies are under evaluation:
-Direct-to-chip cooling, where coolant flows through plates attached directly to server components
-Immersion cooling, which submerges servers in a thermally conductive oil bath
A coolant distribution unit (CDU) plays a central role in maintaining efficient coolant flow and optimizing energy usage.
LG Uplus previewed these solutions earlier at MWC 2025 in Barcelona and the demo room now serves as a real-world testbed. Insights from the trials will inform full-scale liquid cooling adoption at both the Pyeongchon 2 Center—the largest data center in the metropolitan area—and the future Paju AIDC.
The company also plans to launch AIDC tours for customers and industry partners starting later this year.
Jung Sook-kyung, head of LG Uplus’ AIDC business unit, said:“With growing demand for AI and HPC increasing heat loads, advanced cooling is essential. This project marks a milestone in our effort to build sustainable and high-performance AI infrastructure.”
In February, the Korean telco noted that the new liquid cooling solution, developed in collaboration with GST and other partners, uses non-conductive fluid immersion to efficiently dissipate heat from servers.
LG Uplus is also working with partners to develop a Coolant Distribution Unit (CDU) solution, which circulates water to cool server-generated heat.
As artificial intelligence continues to advance at a fast pace, data centers need to evolve to handle the growing demand for computing power. Traditional air-based cooling systems are struggling to keep up with the high heat generated by modern AI hardware. This is where liquid cooling comes in.
Liquid cooling is a method used to remove excess heat from electronic components by using a liquid coolant instead of air. This technique is commonly applied in high-performance computing (HPC), gaming computers and now, AI data centers.
There are two main types of liquid cooling:
-Direct Liquid Cooling (DLC): This method involves circulating liquid directly over hot components, such as processors and GPUs, using small tubes or cold plates. The heat is transferred to the liquid, which then moves it away from the components.
-Immersion Cooling: In this approach, entire servers or components are submerged in a non-conductive cooling liquid. The liquid absorbs the heat and then carries it away through a cooling system.
Both methods of liquid cooling in data centers are far more efficient than traditional air cooling, making them essential for AI workloads.
