A Review of Non-Uniform Load Distribution and Solutions in Data Centers: Micro-Scale Liquid Cooling and Large-Scale Air Cooling

Abstract

Nowadays, the number of transistors on electronic components is increasing exponentially leading to an ultra-high heat flux (106~107 W/m2). The non-uniform load distribution on the chip and the local hot spots in the thermal environment are key issues in a data center (DC). Microchannel liquid cooling is an effective method to inhibit heat accumulation on the chip. Optimizing the air distribution is a crucial approach to realizing energy savings. This study summarizes the latest research on thermal management by microchannel liquid cooling and air distribution optimization in DCs. The existing issues concerning the structure universality of the microchannel heat sink (MCHS), the stability of flow boiling and the new coolant, the prediction of local hot spots, and the intelligent control of the cooling system are identified. Furthermore, a novel cooling strategy of multi-scale synergy is recommended, which is expected to suppress local hot spots and heighten temperature uniformity. This review provides valuable insights into multi-scale thermal management for DCs. It paves the way for applying innovative cooling technologies and artificial intelligence methods to promote the efficient operation and low-carbon retrofit of DCs.

Keywords

• data center
• microchannel liquid cooling
• air distribution
• multi-scale synergy
• intelligent regulation