ARTÍCULO
TITULO

Server Level Liquid Cooling: Do Higher System Temperatures Improve Energy Ef?ciency?

Alexander A. Moskovsky    
Egor A. Druzhinin    
Alexey B. Shmelev    
Vladimir V. Mironov    
Andrey Semin    

Resumen

Liquid cooling is now a mainstream approach to boost energy ef?ciency for high performance computing systems. Higher coolant temperature is usually considered as an advantage, since it allows heat reuse/recuperation and simpli?es datacenter infrastructure by eliminating the need of chiller machine. However, the use of hot coolant imposes high requirements for cooling equipment. A promising approach is to utilize coldplates with channel structure and liquid circulation for heat removal from semiconductor components. We have designed a coldplate with low heat-resistance that ensures effective cooling with only 2030° temperature difference between coolant and electronic parts of a server. Under stress-test conditions the coolant temperature was up to 65 °C while server operation was undisturbed. We also studied power ef?ciency (expressed in ?oating point operations per watt) dependence on the coolant temperature (19-65 °C) on theindividualserverlevel (based on Intel Grantley platform with dual Intel Xeon E5-2697v3 processors). ?The power performance ratio shows moderate (?10%) ef?ciency drop from 19 to 65°C due to increase of leak age current in chipset components and reduction of processor frequency resulted into proportional reduction of DGEMM benchmark performance. It must be taken into account by datacenter designers, that the amount of recuperated energy from 65 °C should be at least?10% to justify the choice of high temperature coolant solution.

 Artículos similares

       
 
Sehrash Ashraf, Shahnaz Parveen Khattak and Mohammad Tariq Iqbal    
Across the globe, COVID-19 had far-reaching impacts that included healthcare facilities, public health, as well as all forms of transport. Hospitals were experiencing staffing shortages at the same time as patients were experiencing healthcare issues. Co... ver más

 
Shuangzhong Wang and Ying Zhang    
The federated learning network requires all the connection weights to be shared among the server and clients during training which increases the risk of data leakage. Meanwhile, the traditional federated learning method has a poor diagnostic effect for n... ver más

 
Nawaf Alharbe, Abeer Aljohani and Mohamed Ali Rakrouki    
Due to the large-scale development of cloud computing, data center electricity energy costs have increased rapidly. Energy saving has become a major research direction of virtual machine placement problems. At the same time, the multi-dimensional resourc... ver más
Revista: Algorithms

 
Qianqian Tong, Guannan Liang, Jiahao Ding, Tan Zhu, Miao Pan and Jinbo Bi    
Regularized sparse learning with the l0 l 0 -norm is important in many areas, including statistical learning and signal processing. Iterative hard thresholding (IHT) methods are the state-of-the-art for nonconvex-constrained sparse learning due to their ... ver más
Revista: Algorithms

 
Jin-young Choi, Minkyoung Cho and Jik-Soo Kim    
Recently, ?Big Data? platform technologies have become crucial for distributed processing of diverse unstructured or semi-structured data as the amount of data generated increases rapidly. In order to effectively manage these Big Data, Cloud Computing ha... ver más
Revista: Applied Sciences