Research On Energy-Efficient Optimization For Operators Composition Of Edge Network In Big Data Environment

With the advent of the Internet of Everything and the era of 5G high-bandwidth and low-latency,the amount of data in industries such as Internet of Vehicles,smart manufacturing,smart cities,and smart homes has exploded,resulting in much more challenges about real-time,low energy consumption and security for computing facilities.How to improve the real-time nature of complex application requests in the network under the background of big data,and effectively save network energy consumption are the current research focus.Edge computing provides intelligent services at the edge of the network close to the data source,which not only improves the response speed of application requests,protects user privacy,but also saves network traffic.As the application requests in the edge network become more and more complicated and diversified,the limited computing,storage and service capabilities of a single smart device are far from meeting the needs of complex applications.How to coordinate multiple operation service nodes at the network edge close to the data source is the focus of this study to complete the massive data processing,analysis and integration,and intelligent prediction.Aiming at how to improve the real-time performance and energy saving of complex application requests in the edge network,the research is carried out from three aspects,constructing the smallest edge cluster for each complex application,encapsulating the feasible solution matrix,and instantiating the operation service combination.The main research contents are as follows:First,according to the data processing requirements on the application request topology tree,the multi-stage shortest path algorithm is used to construct the smallest edge cluster,filter out the unreasonable operation services in the search area,and obtain all feasible solution combinations that meet the application request.Besides,the feasible solutions on the operation service node are also packed.Secondly,we instantiate the operation service nodes included in all feasible solutions.Among them,the constraints of operating service nodes are space-time constraints and energy constraints.Based on the application of topology tree,we analyze the continuous operation service node selection priority strategy.When the operation service nodes are combined on the feasible solution,the three aspects of activation energy consumption,delay energy consumption and transmission energy consumption are analyzed,and an operation service composition problem model is constructed.Then,the instantiated operation service combination is brought into the heuristic algorithm to iterate,and the process of searching the optimal operation combination within the edge cluster range is a multi-objective optimization problem.Finally,experiments are conducted in the edge cluster range and the entire search area,and the experimental results are compared and analyzed in terms of fitness function value,average energy consumption,and minimum residual energy.It is proved that the minimum edge cluster constructed in this paper accelerates the convergence speed of understanding,and the instantiation mechanism of the feasible solution enhances the robustness and stability of the gray wolf optimization algorithm,and meets the real-time and energy-saving goals of request response.