GlobalData’s latest Strategic Intelligence report, “Deep Dive into The Environmental Impact of Data Centers,” reveals that water is the most common cooling method, and its use has increased during the current AI boom. However, new cooling technologies are being developed.
Martina Raveni, Analyst in the Strategic team at GlobalData, comments: “The high temperature of data centers is a critical issue at the moment. If equipment overheats, malfunctions and breakdowns can occur, with repercussions for the many sectors that rely on those data centers. As demand for AI applications increases, managing these temperatures will become increasingly important.”
Big Tech companies training large language models (LLMs) are increasing their water consumption for cooling purposes. However, water cooling is not ideal in the long-term due to water scarcity in certain areas and because it is ultimately a finite resource.
To reduce water consumption, new cooling technologies are being developed, including immersion cooling using dielectric fluids and ceramic components for circuit boards and semiconductors. Future cooling methods such as underwater data centers and data centers in space are also being explored.
Raveni concludes: “There will likely be increased deployment of data centers in cold climates. Expectations include greater adoption of sustainable design and advanced cooling technologies like immersion cooling, driven by data centers’ energy efficiency demands and regulatory pressures.
“The future of data centers hinges on our ability to innovate and adapt to the pressing environmental challenges they pose. As we embrace advanced cooling technologies and sustainable practices, it is crucial to balance the growing demand for data processing with our responsibility to protect finite resources. The transition to more efficient and eco-friendly solutions will not only enhance operational resilience but also pave the way for a more sustainable digital infrastructure.”