Data center liquid cooling market to grow amid AI boom
The market for data center liquid cooling is projected to grow at a compound annual growth rate of 28.7%, driven by widespread AI adoption and increasing GPU thermal loads. The shift is also supported by corporate sustainability targets and a move toward liquid-first data center designs. This trend reflects the growing infrastructure demands of large-scale AI and machine learning workloads.
- Liquid cooling is significantly more effective at heat transfer than traditional air cooling, with some estimates suggesting it's up to 1,000 times more efficient. This allows data centers to achieve a much lower Power Usage Effectiveness (PUE), a metric for energy efficiency. Liquid-cooled facilities can reach PUE ratios as low as 1.05-1.15, compared to 1.4-1.8 for air-cooled data centers. - The latest generation of AI-powering GPUs from manufacturers like Nvidia have a thermal design power exceeding 700W, with some newer models consuming up to 1400W, making liquid cooling a necessity. Traditional air cooling struggles to manage the heat generated by these high-density racks, which can exceed 50 kW and are projected to reach 200-250 kW in the future. - The global market for data center liquid cooling was valued at approximately $4.8 billion in 2025 and is projected to reach over $27 billion by 2035. Other forecasts project the market to expand from $6.6 billion in 2026 to $38.4 billion by 2033. - Two primary liquid cooling technologies are gaining traction: direct-to-chip (D2C) and immersion cooling. D2C systems circulate liquid through cold plates attached directly to processors and are easier to integrate into existing rack setups. Immersion cooling submerges entire servers in a non-conductive fluid, offering maximum heat dissipation and is considered the fastest-growing segment. - Key players in the liquid cooling market include established companies like Schneider Electric, Vertiv, and Rittal, alongside specialists such as CoolIT Systems, Green Revolution Cooling (GRC), and LiquidStack. These companies offer a range of solutions from direct-to-chip cold plates to full immersion systems. - Hyperscalers like Meta, Microsoft, and Google are already making major deployments of liquid cooling to support their AI infrastructure. For example, Microsoft's Azure regions utilize liquid cooling to support OpenAI's GPT models. - North America currently holds the largest share of the data center liquid cooling market, accounting for over a third of the revenue in 2025. This is driven by the high concentration of hyperscale cloud providers and AI companies in the region. - While highly efficient, liquid cooling systems have higher upfront costs compared to traditional air cooling. However, the operational savings from reduced energy consumption can lead to a return on investment within 2-4 years for many deployments.
