Abstract

The continuous growth of data centers in recent years has made them significant power consumers, reaching about 2% of the global electricity demand. About half of the energy consumption in a typical data center is due to cooling systems required to keep the components within specific temperature limits. Furthermore, most of the energy supplied to the IT components is converted into waste heat. Therefore, it is important to investigate possible solutions to optimize the thermal management and the potential for waste heat recovery. The present paper reports an optimization study carried out in order to achieve energy savings and improve heat recovery potential for a hybrid air-liquid cooled server. The effect of passive (baffle) and active (synthetic jet) flow control systems is investigated. Entropy generation minimization (EGM) is employed to evaluate the cooling performance in the server. In addition, to enhance the potential for waste heat recovery, the average outlet temperature and temperature uniformity along the server outlet are maximized. Promising results in terms of cooling efficiency and waste heat recovery potential are achieved, i.e. the required inlet velocity from the fans is reduced up to 50%, the total entropy generation rate in the server is reduced up to 40%, the average outlet temperature is enhanced up to 7 K, with improved uniformity up to 60%.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call