To reduce data center energy consumption, this study proposes a novel multi-timescale scheduling model based on load characteristics. In the day-ahead phase, by analyzing the underlying logical characteristics of batch workloads—including arrival rate, delay tolerance duration, execution time, and internal DAG (Directed Acyclic Graph) structure—a matrix model is established to participate in scheduling optimization, thereby smoothing server energy consumption curves. In the intraday phase, a rolling optimization model is constructed to dynamically adjust intraday plans by integrating redundancy backup and geo-migration strategies to address load fluctuations. The model is validated using the YALMIP + CPLEX solver. Results demonstrate that the proposed model significantly reduces the operational costs of data centers and offers substantial practical value for enhancing energy efficiency management capabilities.
WANG P,CAO Y J,DING Z H,et al.Stochastic programming for cost optimization in geographically distributed internet data centers[J].CSEE Journal of Power and Energy Systems,2022,8(4):1215-1232.
XU R,MITRA S,RAHMAN J,et al.Pythia:improving datacenter utilization via precise contention prediction for multiple co-located workloads[C]// Proceedings of the19th International Middleware Conference.New York,USA,2018:146-160.
SUN Q H,WU C,LI Z P,et al.Colocation demand response: Joint online mechanisms for individual utility and social welfare maximization[J].IEEE Journal on Selected Areas in Communications,2016,34(12):3978-3992.
CHEN W,YE K,WANG Y,et al.How does theworkload look like in production cloud? Analysis and clustering of workloads on Alibaba cluster trace[C]// Proceedings of the 25th International Conference on Parallel and Distributed Systems(ICPADS 2019).Piscataway,USA,2019:102-109.
GU Z C,TANG S H,JIANG B L,et al.Characterizing job-task dependency in cloud workloads using graph learning[C]//Proceedings of the 2021 IEEE International Parallel and Distributed Processing Symposium Workshops(IPDPSW).Piscataway,USA,2021:288-297.
JIANG C F,HAN G J,LIN J B,et al.Characteristics of co-allocated online services and batch jobs in internet data centers:A case study from Alibaba Cloud[J].IEEE Access,2019,7: 22495-22508
YU J Y,TONG W,LV P Z,et al.TERMS: taskmanagement policies to achieve high performance for mixed workloads using surplusresources[J].Journal of Parallel and Distributed Computing,2022,170:74-85.
JIANG C F,QIU Y T,SHI W S,et al.Characterizing co-located workloads in Alibaba cloud datacenters[J].IEEE Transactions on Cloud Computing,2022,10(4):2381-2397.
DING Z H,CAO Y J,XIE L Y,et al.Integrated stochastic energy management for datacenter microgrid considering waste heat recovery[J].IEEE Transactions on IndustryApplications,2019,55(3): 2198-2207.