基于服务的云边端协同流数据处理体系结构研究  

Architecture for Stream Processing with Service-based Method in Cloud-Edge Collaboration Environment

作　　者：张守利 刘晨[2] ZHANG Shou-li;LIU Chen(College of Information Science and Engineering/Shandong Agricultural University,Tai'an 271018,China;Beijing Key Laboratory on Integration and Analysis of Large-Scale Stream Data,North China University of Technology,Beijing 100049,China)

机构地区：[1]山东农业大学信息科学与工程学院,山东泰安271018 [2]北方工业大学大规模流数据集成与分析技术北京市重点实验室,北京100144

出　　处：《山东农业大学学报（自然科学版）》2024年第3期385-395,共11页Journal of Shandong Agricultural University：Natural Science Edition

摘　　要：物联网相关技术的快速发展产生了大规模传感流数据和对流数据的高并发处理需求,云边端协同计算正成为低延迟、高可靠的流数据处理的有效途径。为了提升流数据处理系统的灵活性和可扩展性,降低流数据处理延迟,本文提出一种基于服务的分散式云边端协同流数据处理体系结构,设计了面向大规模流数据的主动式数据服务模型,流数据及流数据处理被抽象为合适粒度、可被独立部署和动态调度的服务,解耦数据与计算。引入事件驱动机制,提出了基于事件驱动的云边端服务动态协作机制,有效提升了系统的灵活性。基于真实的电能质量传感流数据验证了本文所提出架构的正确性和有效性。The rapid development of Internet of Things technologies has generated large-scale sensor streaming data and high-concurrency processing requirements for streaming data.The cloud-edge collaborative computing is becoming an effective approach for low-latency,highly reliable stream data processing.In order to improve the flexibility and scalability of stream data processing system,reduce the delay of stream data processing,this paper proposes a decentralized architecture for streaming data processing with a service-based method in cloud-edge collaborative environment.A proactive data service model for large-scale streaming data services is designed.The streaming data and streaming data processing are abstracted into appropriate grained services that can be independently deployed and dynamically scheduled,decoupling data from computation.By introducing the event-driven mechanism,we proposed the event-driven based cloud edge service dynamic collaboration mechanism to improve the flexibility of the system effectively.A simulation-based evaluation based on realworld power quality sensor streaming data verified the effectiveness and efficiency of our approach.

关 键 词：流数据处理 云边端协同 服务计算 事件驱动 服务协作 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]