Impact of coolant distribution design on server-level thermal management in data centers

M. Khoshvaght-Aliabadi, P. Ghodrati, J.Y. Shin and Y.T. Kang

Energy, 2025, vol. 330, issue C

Abstract: A significant portion of data center servers (DCSs) energy is used for CPU cooling, necessitating efficient coolant distribution to optimize thermal management and improve energy efficiency. This study investigates and analyzes various coolant distribution designs for jetted cold plate cooling, with a particular focus on DCSs configured with two and four CPUs. The BBD-RSM approach effectively determines the key design parameters for optimizing cold plate configurations. In addition, the analysis provides guidance on selecting an appropriate coolant mass flow rate to enhance thermal performance and energy efficiency in CPU cooling applications. In the two-CPU case, the design with a smooth cold plate on the first CPU and a finned cold plate on the second achieves the lowest thermal resistance, improving temperature uniformity by 23.5 % and 78.5 % compared to the series and parallel arrangements, respectively. For the four-CPU case, a series design is recommended to achieve lower thermal resistance and improved temperature uniformity, while a parallel design is preferred to minimize pumping power consumption. However, a hybrid configuration that strategically uses finned cold plates results in the optimal design, reducing pumping power consumption by at least 10 % and limiting the temperature difference between the CPUs to approximately 1 K.

Keywords: Data center; Thermal management; Server-level; CPU cooling; Cold plate; Optimization (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0360544225025782
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:energy:v:330:y:2025:i:c:s0360544225025782

DOI: 10.1016/j.energy.2025.136936

Access Statistics for this article

Energy is currently edited by Henrik Lund and Mark J. Kaiser

More articles in Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().