Abstract
The flow distribution and central processing unit (CPU) temperatures inside a rack of thirty 1 U (single rack unit) Sun Fire V20z servers retrofitted with direct-to-chip liquid cooling and two coolant pumping configuration scenarios (central and distributed) are investigated using the EPANET open source network flow software. The results revealed that the servers in the top of the rack and close to the cooling distribution unit can receive up 30% higher flow rate than the servers in the bottom of the rack, depending on the pumping scenario. This results in a variation in the CPU temperatures depending on the position in the rack. Optimization analysis is carried out and shows that increasing the flow distribution manifold’s dimensions can reduce the flow variation through the servers and increase the total coolant flow rate in the rack by roughly 10%. In addition, activating the small pumps in the direct-to-chip liquid cooling loops inside the servers (distributed pumping) resulted in an increase of 2 °C in the CPU temperatures at the high computational workload.
Acknowledgments
The authors would like to thank Airedale International Air Conditioning Ltd., for the collaboration and providing the dry air cooler modified with a Carel spray system. We would like to acknowledge the Advanced Research Computing unit from the University of Leeds for donating the Sunfire V20z systems. We would like to thank CoolT systems for providing the liquid cooling systems used.
Additional information
Funding
Notes on contributors
![](/cms/asset/6cc82984-6a20-4d19-9327-bcc04642a7a5/uhte_a_1640467_ilg0001_c.jpg)
Mustafa A. Kadhim
Mustafa Kadhim received his PhD in Thermofluids, specializing in data center liquid cooling from the University of Leeds (2018). His project was to engineer high-performance operation of the information technology equipment. This is part of his general interest of becoming a world-leading expert in thermal management of data centers. His current career is in industry to engineer data centers toward lower power consumption, higher computational performance, and energy reuse.
![](/cms/asset/f2441aa6-8087-4903-a2c8-668313f61dff/uhte_a_1640467_ilg0002_c.jpg)
Nikil Kapur
Nikil Kapur is a Professor of Applied Fluid Mechanics at the School of Mechanical Engineering, University of Leeds. His research interests span fundamental fluid mechanics phenomena covering multiphasic systems through to industrially related studies. His work is highly disciplinary covering areas as diverse as cell growth, protein aggregation, the design of flow reactors, crystallization, and the design of flow equipment. He draws on computational and experimental techniques to further the understanding of fluid mechanics-related challenges.
![](/cms/asset/41a0ae0b-a415-4c14-87b2-9282f756d512/uhte_a_1640467_ilg0003_c.jpg)
Jonathan L. Summers
Jon Summers is currently on a study leave from the School of Mechanical Engineering at the University of Leeds in the role of Scientific Leader in Data Centers at Research Institutes of Sweden and an Adjunct Professor in Fluid Mechanics at Lulea University of Technology. He has focused his attention toward holistic data center research to bridge the gap between the facility and the information technology, to enable innovative integrated approaches against the backdrop of increasing energy demand. During the last 25 years, he has worked on many government- and industry-funded research and development projects, many of which in the last 8 years have been directed toward energy efficient thermal management of data centers and microelectronics.
![](/cms/asset/6ad03143-d643-4df1-b0e2-0eacaaf63b68/uhte_a_1640467_ilg0004_c.jpg)
Harvey Thompson
Harvey Thompson is a Professor of Computational Fluid Dynamics (CFD) and Head of the School of Mechanical Engineering at the University of Leeds. He has published over 170 papers in the areas of CFD, heat transfer, and multidisciplinary design optimization of engineering products and processes and has collaborated with leading companies in the aerospace, automotive, and food sectors. His recent research in CFD-enabled flow optimization has been focused in electronics and machine tool cooling systems and is finding application in a range of other biological, chemical, and nuclear flow processing systems.