Free Cooling in High Heat Density Data Centers
Chang Ge, Xian Qing Zheng
Abstract
Chang Ge, Xian Qing Zheng
Abstract
Data centers comprised of racks and corresponding servers form the backbone of today’s cloud computing. Continuous operation of large numbers of servers can generate large amounts of heat which in turn requires high capacity cooling systems. These cooling systems can consume a significant portion of the energy required to run a data center and can negatively impact data center efficiency. With increasing operational costs, the data center industry has started to actively search for efficiency improvements to slow down the rising cost of running services. One of the most prominent ways to cut power consumption is free cooling, which uses outside air for cooling IT equipment completely or part of the time. This system is often used in colder climates and requires less energy since it doesn’t use compressors for cooling incoming air. This paper presents an evaluation of Shanghai as a data center location. As temperature affects data center cooling consumption, focus of the paper is on evaluating optimal operating conditions for local data centers.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Data centers comprised of racks and corresponding servers form the backbone of today’s cloud computing. Continuous operation of large numbers of servers can generate large amounts of heat which in turn requires high capacity cooling systems. These cooling systems can consume a significant portion of the energy required to run a data center and can negatively impact data center efficiency. With increasing operational costs, the data center industry has started to actively search for efficiency improvements to slow down the rising cost of running services. One of the most prominent ways to cut power consumption is free cooling, which uses outside air for cooling IT equipment completely or part of the time. This system is often used in colder climates and requires less energy since it doesn’t use compressors for cooling incoming air. This paper presents an evaluation of Shanghai as a data center location. As temperature affects data center cooling consumption, focus of the paper is on evaluating optimal operating conditions for local data centers.
Key concepts: Data center, Free cooling, Server, Center (category theory), Cloud computing, Energy consumption, Water cooling, Cooling capacity