基于边云协同的流数据处理任务卸载优化研究  

Research on Offloading Optimization of Stream Data Processing Tasks Based on Edge-cloud Collaboration

作　　者：徐扬 张萌萌[2] XU Yang;ZHANG Meng-meng(Tianjin Yihualu Information Technology Co.,Ltd.,Tianjin 300350;College of Information Science,North China University of Technology,Beijing 100144)

机构地区：[1]天津易华录信息技术有限公司,天津300350 [2]北方工业大学信息学院,北京100144

出　　处：《数字技术与应用》2021年第7期80-82,共3页Digital Technology & Application

摘　　要：近年来,物联网技术发展迅速,网络边缘产生了大量数据,云计算模式难保障数据处理实时性。为此本文基于边云协同技术对物联网环境下流数据处理任务执行优化展开了研究。本文分析了基于边云协同的流数据处理架构,针对如何减小边云传输的数据量,提出了基于FF(Ford-Fulkerson)算法的任务卸载方法。该算法主要思想是建立基于"流图"的流数据处理任务的资源模型,通过最小割算法对流图进行切割,提取边缘执行任务子图及云端任务子图,减少处理任务宗的数据传输量从而降低服务时延。最后与CloudMethod算法进行了对比,仿真结果表明,这种方法具有一定的有效性。In recent years,the Internet of Things technology has developed rapidly,a large amount of data is generated at the edge of the network,and the cloud computing model is difficult to guarantee the real-time data processing.For this reason,this paper has carried out research on the optimization of the execution of stream data processing tasks in the Internet of Things environment based on the edge-cloud collaboration technology.This paper analyzes the stream data processing architecture based on edge-cloud collaboration,and proposes a task offloading method based on the FF(Ford-Fulkerson)algorithm for how to reduce the amount of data transmitted by the edge-cloud.The main idea of the algorithm is to establish a resource model for stream data processing tasks based on"flow graphs",cut the flow graphs through the minimum cut algorithm,extract edge execution task subgraphs and cloud task subgraphs,and reduce the amount of data transmission for processing tasks.Reduce service delay.Finally,it is compared with the CloudMethod algorithm,and the simulation results show that this method has a certain effectiveness.

关 键 词：边缘计算 边云协同 流数据处理 任务卸载 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]