Performance evaluation and optimization of data center servers using single-phase immersion cooling

Huijuan Wang,Xuejun Yuan,Kun Zhang,Xujin Lang,Hua Chen,Huimin Yu,Shengtao Li
DOI: https://doi.org/10.1016/j.ijheatmasstransfer.2023.125057
IF: 5.2
2023-12-17
International Journal of Heat and Mass Transfer
Abstract:This paper conducts a comprehensive performance evaluation and optimization of single-phase immersion cooling (SPIC) systems to address challenges in data center server cooling. A three-dimensional model of blade servers with different cooling methods is constructed and numerically solved to compare thermal transport performance, flow characteristics, and energy consumption performance. Furthermore, the influence mechanisms of different oil coolants and operating conditions on SPIC system performance are studied. The results indicate that the SPIC scheme exhibits significant advantages in terms of heat transfer capability, heat transfer uniformity, and flow resistance loss. Moreover, the power usage effectiveness (PUE) of SPIC systems decreases by 20.8 % and 17.6 % compared to forced air cooling and water cooling plate solutions, respectively. The dynamic viscosity of oil coolants has the greatest impact on the SPIC performance. Moreover, high-viscosity silicone oil increases the flow resistance of SPIC systems and weakens the convective heat transfer capability between the coolant and components. By ensuring safe operation, increasing the inlet temperature of oil coolants reduces flow resistance losses and enhances heat transfer capacity. Additionally, initially increasing the mass flow rate improves convective heat transfer capacity. However, the heat transfer improvement gradually diminished beyond a certain threshold while flow resistance losses increased significantly.
engineering, mechanical,thermodynamics,mechanics
What problem does this paper attempt to address?