孙毅, 陈恺, 左强, 卢达, 杨泓玥, 李跃. 考虑5G通信负荷协同优化的云边计算网络能量管理模型[J]. 中国电机工程学报, 2023, 43(23): 9020-9032. DOI: 10.13334/j.0258-8013.pcsee.221697
引用本文: 孙毅, 陈恺, 左强, 卢达, 杨泓玥, 李跃. 考虑5G通信负荷协同优化的云边计算网络能量管理模型[J]. 中国电机工程学报, 2023, 43(23): 9020-9032. DOI: 10.13334/j.0258-8013.pcsee.221697
SUN Yi, CHEN Kai, ZUO Qiang, LU Da, YANG Hongyue, LI Yue. Energy Management Model of Cloud-edge Computing Network Considering the Coordinated Optimization of 5G Communication Load[J]. Proceedings of the CSEE, 2023, 43(23): 9020-9032. DOI: 10.13334/j.0258-8013.pcsee.221697
Citation: SUN Yi, CHEN Kai, ZUO Qiang, LU Da, YANG Hongyue, LI Yue. Energy Management Model of Cloud-edge Computing Network Considering the Coordinated Optimization of 5G Communication Load[J]. Proceedings of the CSEE, 2023, 43(23): 9020-9032. DOI: 10.13334/j.0258-8013.pcsee.221697

考虑5G通信负荷协同优化的云边计算网络能量管理模型

Energy Management Model of Cloud-edge Computing Network Considering the Coordinated Optimization of 5G Communication Load

  • 摘要: 随着物联网发展,数据中心与5G网络相继成为可观的用能负荷,给低碳网络建设带来挑战。降低计算网络对电网负担,提升可再生能源应用比例并削减购电成本是实现低碳计算目标面临的重要议题。该文首先建立考虑电网、可再生能源混合供能的跨域数据中心与5G边缘计算网络用能调度模型;基于信息−能源流耦合性,提出联合任务卸载、功率控制与云边计算负载迁移策略,实现网络购电成本的空间转移、削减;其次,考虑时延、服务机制与能耗的耦合性,设计基于主从博弈的云计算服务定价机制,鼓励计算资源共享实现云边用能行为优化。仿真分析表明,所提方法能够提高云边网络合作积极性,有效减少各方网络主体的购电成本并降低电网负载压力。

     

    Abstract: With the development of Internet of Things, data center and 5G network have become heavy load in succession, and bring challenges to the construction of low-carbon network. Reducing the burden of computing network to power grid, improving the proportion of renewable energy consumption and cutting down the electricity purchasing cost are important issues to achieve low-carbon computing. This paper firstly investigate a load dispatch model considering grid, renewable energy, data center and 5G network interaction, and based on the coupling of data-energy flow, a joint task offloading, power control and computing load migration strategy is proposed to realize the transfer and reduction of network energy cost. Then, by considering the coupling among energy, service mechanism and delay, a Stackelberg game-based computing pricing mechanism is studied to encourage the resource sharing between 5G network and data center to reduce the energy cost conjointly. Simulation results show that the proposed strategy can improve the cooperation positivity, reduce the electricity purchasing cost, and lower the burden on power grid efficiently.

     

/

返回文章
返回