A comprehensive review of a data centre for a cooling system

• Main Authors: Gerijih, Debie Devisser, Hazrati, K. Z., Qibtiah, Raja Mariatul, Baharudin, N., Razali, Nizamuddin, Ali, Khairul Fahmi
• Format: Article
• Language: English
• Subjects: T Technology (General)
• Online Access: http://eprints.uthm.edu.my/7185/1/J13996_53bd9f016f714f7cae106d14fd3ad5b1.pdf; http://eprints.uthm.edu.my/7185/

Cyber-Physical-Social Systems, commercial enterprises, and social networking use data centers to store, process, and distribute massive amounts of data. A data center serves as the foundation for all of these endeavors. The data center's workload and power consumption are increasing rapidly due to the demand for remote data services. Mechanical refrigeration and terminal cooling are the most critical components for most cooling systems. There is a way to transfer heat from the data center to the outside environment, but it's a complicated process. Air cooling systems and technology are most useful for room cooling and rack-level cooling. Because of their superior cooling performance and higher energy efficiency, air cooling has attracted more attention than water cooling in most existing data centers. The chillers and fans consume the most power of all the cooling equipment in the system. These methods can be divided into mechanism-based methods and data-driven methods for energy management of the cooling system. Operation management of cooling equipment is proposed to reduce power consumption, mainly using predictive model control and reinforcement learning-based methods. An overview of the data center's cooling system is presented in this paper, which focuses on the most common cooling solutions, power consumption modeling methods, and optimization control strategies, among others. In addition, the data center's cooling system is described as a current and future issue.