面向边缘光算力网络的上行链路资源协同调度算法  

Uplink resource coordinated scheduling algorithm for edge-oriented optical computing power networks

作　　者：王蕴 林霄 楼芝兰 李军[3] 孙卫强[4] WANG Yun;LIN Xiao;LOU Zhilan;LI Jun;SUN Weiqiang(College of Physics and Information Engineering,Fuzhou University,Fuzhou 350116,China;School of Data Sciences,Zhejiang University of Finance&Economics,Hangzhou 350018,China;Jiangsu New Optical Fiber Technology and Communication Network Engineering Research Center,Soochow University,Suzhou Jiangsu 215006,China;State Key Laboratory of Advanced Optical Communication Systems and Networks,Shanghai Jiao Tong University,Shanghai 200240,China)

机构地区：[1]福州大学物理与信息工程学院,福州350116 [2]浙江财经大学数据科学学院,杭州310018 [3]苏州大学江苏省新型光纤技术与通信网络工程研究中心,江苏苏州215006 [4]上海交通大学区域光纤通信网与新型光通信系统国家重点实验室,上海200240

出　　处：《光通信技术》2024年第3期45-51,共7页Optical Communication Technology

基　　金：国家自然科学基金青年项目(批准号:61901118、12001483)资助;上海交通大学“区域光纤通信网与新型光通信系统国家重点实验室”开放基金(批准号:2023GZKF020)资助;江苏省新型光纤技术与通信网络工程研究中心开放研究课题(批准号:SDGC2232)资助。

摘　　要：为满足冷、热业务实时、高效的算力调度需求,提出一种基于自适应噪声完全集合经验模态分解(CEEMDAN)与时间卷积网络(TCN)的算力负载预测模型(简称C-TCN模型),并设计了基于C-TCN与Q学习的资源协同调度算法(CTQ算法),利用C-TCN模型提前感知下一时刻负载变化,通过Q学习协同调度波长与存储资源,寻找最佳波长划分与边缘存储分配方案。实验结果表明:CTQ算法的调度性能不仅优于现有调度算法,能满足冷、热业务调度性能要求,而且还能提高波长利用率。In order to meet the real-time and efficient computing power scheduling requirements of hot and cold services,a com-putational load prediction model(abbreviated as C-TCN model)based on adaptive noise complete set empirical mode decompo-sition(CEEMDAN)and time convolutional network(TCN)is proposed,and a resource cooperative scheduling algorithm(CTQ algorithm)based on C-TCN and Q learning is designed.The C-TCN model is used to sense the load change at the next time in advance,and the optimal wavelength partitioning and edge storage allocation scheme is found through Q learning.The experi-mental results show that the CTQ algorithm not only has better scheduling performance than the existing scheduling algorithms,but also can meet the requirements of hot and cold service scheduling performance,and improve the wavelength utilization rate.

关 键 词：边缘光算力网络 算力调度 数据传输 资源调度 网络优化 

分 类 号：TN91[电子电信—通信与信息系统]