Abstract
Increasing power densities in data centers due to the rise of artificial intelligence, high-performance computing, and machine learning compel engineers to develop new cooling strategies and designs for high-performance information technology (IT) equipment. Two-phase cooling is a promising technology that exploits the latent heat of the coolant which is significantly more effective in removing high heat fluxes than when using the sensible heat of the fluid. Also, utilizing the latent heat allows operating at lower coolant flow rates and implies more uniformity in the temperature of heated surfaces. Despite the benefits of two-phase cooling, the phase change adds complexities to a system when multiple evaporators (exposed to different heat fluxes potentially) are connected to a single coolant distribution unit. In this article, a commercial coolant distribution unit is used to investigate pumped two-phase cooling in rack scale. Seventeen two-rack unit servers from two distinct models are retrofitted with 34 impinging jet evaporators and deployed in a rack. Four case studies are presented to provide insights into the complex behavior of a pumped two-phase cooling system with several evaporators. The flow rates and pressure distribution across the rack are studied in various filling ratios. Also, investigated is the transient behavior of the cooling system due to a step change in the IT workload. Finally, a control system is designed to regulate the temperature of the supplied coolant in response to the step change in the IT workload and is tested.