Research On Edge Information System Architecture Design And Resource Management

Under the background of new intelligent and networked battlefields,the form of warfare has gradually evolved from traditional“large-scale conventional wars” and “smallscale unconventional wars” to “hybrid wars” with more integrated combat modes.Hybrid wars are highly uncertain,high-frequency,and difficult to predict.In this context,edge command and control(EC2)emerges as a new concept.With the edge networks,edge computing,edge intelligence,and other new-generation technologies,EC2 aims to perceive large-scale battlefield environments and enable the networked and decentralized edge through extensive information sharing and team collaboration.In this way,the edge can be self-task,self-action,self-organization,and highly adaptive.The current integrated information infrastructure is mainly built on the cloud data center,which has serious problems of lagging response in resource management and service distribution when dealing with big data and providing combat services.It cannot support the paradigm of edge command and control and cannot meet real-time service requirements of the evolving battlefields.In this case,the edge computing architecture for big data applications is an important way to break through the limitations of the current cloud-based architecture and improve the network service quality and resource utilization.Specifically,this paper mainly focuses on two key issues of edge computing architecture: 1)system architecture design and 2)resource management,and the detailed as follows.1)In the architecture design part,to solve the resource and service interoperability problem between edge information infrastructure of different categories of troops,this paper proposes a resource management and service distribution architecture with dynamic resource allocation.This architecture can transform the former independent information environment into the networked,so as to seamlessly realize the resource cooperation and service provisioning across standalone edge infrastructure providers(EIPs)and clouds.To efficiently schedule and utilize the resources across multiple EIPs,we systematically characterize the provisioning process as a large-scale linear programming problem and transform it into an easily solved form.Accordingly,we design a dynamic algorithm to tackle the varying service demands from users.Extensive experiments can show the effectiveness of edge federation.2)Based on the architecture,we present three works in the resource management part.At the central layer of the edge information infrastructure,this paper proposes the Interactive Temporal Recurrent Convolution Network(ITRCN)to predict interactive network traffic in the large-scale data center network.More specifically,ITRCN takes communications between services as a whole and directly predict interactive traffic.Within the ITRCN model,the Convolution Neural Network(CNN)part learns network traffic as images to capture the network-wide services’ correlations,the GRU part learns the temporal features to help the interactive network traffic prediction.We conducted comprehensive experiments based on real-world network traffic datasets,and the results show the superiority of ITRCN.3)At the edge layer of the edge information infrastructure,in order to improve the collaboration of heterogeneous edge devices,we investigate the worker selection problem in vehicle-based crowdsourcing.We first conduct a comprehensive data analytics on two real-world vehicle traces and obtain instructive data insights.Inspired by these insights,we propose the Performance transfer based Online worker SElection(POSE)scheme,which works independently from trajectory prediction with two components,i.e.,transfer learning based performance estimation and online worker selection.Based on the trajectory pattern,the former component collects a short-period trajectory penetration data of vehicles for model fitting,which can output a specific numerical distribution.With the fitting model,we can identify and select vehicles with high trajectory penetration at the initial stage to cope with the “cold start” problem.Then,we map the worker selection problem into a multi-armed bandit problem and develop the upper confidence boundbased approach to solve it.Extensive trace-driven simulations are carried out and the results demonstrate the efficiency of POSE in terms of cumulative platform utility.4)At the edge layer of the edge information infrastructure,in order to optimize resource utilization,we investigate the edge cache deployment for high definition(HD)urban map provisioning,which is an essential building block for future autonomous driving.Given a deployment budget,we first formulate a satisfied downloading request maximization(SDRM)problem to obtain the deployment locations and customized cache sizes.The SDRM problem is unsolvable directly as the urban transportation traffic is highly dynamic and future traffic conditions are unknown in advance.Based on data analytics of the vehicular GPS trace,we propose an architecture named Co De,built on which we transform and address the SDRM problem.The reference implementation of Co De is respectively developed based on the insights from two urban-scale carpool GPS traces.The novelty and contributions of Co De lie in its three-layer design.Particularly,at data feeding layer,we make use of two urban 60-day GPS traces involving respectively 37,801 and17,517 vehicles,to extract the analytics samples.At mobility characterization layer,we conduct extensive data analytics on traffic mobility in terms of their distribution,correlation,and variation,to mine the crucial traffic mobility patterns for strategy customization.At cache deployment layer,we propose the K-order subgraph for each block to record the moving statistics within its K-order neighborhood,and transform the SDRM problem accordingly.Then,the Route we Ight ba Sed gr Eedy(RISE)algorithm is devised for the problem,which can deliver the deployment decisions.Extensive data-driven experiments are carried out to demonstrate the superior performance of Co De in terms of request hit ratio and caching resource utility.Under the edge combat scenario,the research of architecture design and resource management in this paper can significantly promote the construction and development of edge information architecture and has critical scientific and military value.