TY - GEN
T1 - Comparison of Deep Reinforcement Learning Algorithms in Data Center Cooling Management
T2 - 2021 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2021
AU - Hua, Tianyang
AU - Wan, Jianxiong
AU - Jaffry, Shan
AU - Rasheed, Zeeshan
AU - Li, Leixiao
AU - Ma, Zhiqiang
N1 - Funding Information:
This work was funded in part by the National Natural Science Foundation of China (NSFC) under Grants. 61862048, 61762070, and 61962045, Inner Mongolia Key Technological Development Program (2019ZD015), Inner Mongolia Autonomous Region Special Program for Engineering Application of Scientific and Technical Payoffs (2021CG0033, 2020CG0073), and Key Scientific and Technological Research Program of Inner Mongolia Autonomous Region (2019GG273).
Publisher Copyright:
© 2021 IEEE.
PY - 2021
Y1 - 2021
N2 - The growth in scale and power density of Data Centers (DC) poses serious challenges to the cooling management. Recently, there are many studies using machine learning to solve the cooling management problems. However, a comprehensive comparative study is still missing. In this work, we compare the performance of various Deep Reinforcement Learning (DRL) algorithms, including Deep-Q Networks (DQN), Deep Deterministic Policy Gradient (DDPG), and Branching Dueling Q-Network (BDQ), using the Active Ventilation Tiles (AVTs) control problem in raised-floor DC as an example. In particular, we design two multiagent algorithms based on DQN and three critic architectures for DDPG. Simulations based on real world workload show that DDPG provides the best performance over the considered algorithms.
AB - The growth in scale and power density of Data Centers (DC) poses serious challenges to the cooling management. Recently, there are many studies using machine learning to solve the cooling management problems. However, a comprehensive comparative study is still missing. In this work, we compare the performance of various Deep Reinforcement Learning (DRL) algorithms, including Deep-Q Networks (DQN), Deep Deterministic Policy Gradient (DDPG), and Branching Dueling Q-Network (BDQ), using the Active Ventilation Tiles (AVTs) control problem in raised-floor DC as an example. In particular, we design two multiagent algorithms based on DQN and three critic architectures for DDPG. Simulations based on real world workload show that DDPG provides the best performance over the considered algorithms.
UR - http://www.scopus.com/inward/record.url?scp=85124301168&partnerID=8YFLogxK
U2 - 10.1109/SMC52423.2021.9659100
DO - 10.1109/SMC52423.2021.9659100
M3 - Conference Proceeding
AN - SCOPUS:85124301168
T3 - Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics
SP - 392
EP - 397
BT - 2021 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2021
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 17 October 2021 through 20 October 2021
ER -