Research On Optimization Of Data Processing Delay And Energy Consumption In Fog Computing

With the rapid development of Internet of Things(IoT)technology,the number of mobile terminal devices is increasing,which will produce massive data.How to compute,store and manage these data safely and efficiently is an urgent problem to be solved.Cloud computing as a centralized processing technology,inputs all the data into cloud servers for processing,which will bring two problems.First,the end users transmit a large amount of data to the cloud,which will not only consume a lot of energy,time and bandwidth resources,but also cause tremendous pressure on the transmission link.Second,cloud servers cover a wide range and the users are far away from them,which will lead to a large transmission delay.Fog computing extends cloud computing services to the edge of the network to solve the problems in cloud computing through geographical distribution,location sensing,mobility support and so on.Therefore,the thesis selects the optimization of data processing delay and energy consumption in fog computing as the research topic.In this thesis,we mainly study the problem of delay and energy consumption of data processing in fog computing.Combined with a new fog computing framework,the problem was analyzed and validated by establishing mathematical model and designing algorithm.The specific research contents are as follows.(1)Optimization algorithm design of data processing delay in fog computing.Based on the cloud computing architecture,we study the data processing delay problem combined with three-layer network architecture model which combines cloud computing and fog computing(CFNAM).In fog computing layer,we abstract the network topology graph of fog devices into weighted undirected graph and propose a computational scheme of mutual cooperation between the fog devices.Then,we use the Kruskal algorithm to compute the minimum spanning tree of weighted undirected graph,so as to reduce the communication delay between them.Besides,we divide and assign the tasks based on the constrained optimization problem,and solve the computation delay of fog nodes by the Lagrange multiplier method.In cloud computing layer,we focus on the balanced transmission method to solve the data transmission delay from fog devices to cloud servers,and obtain an optimal allocation matrix,which reduces the data communication delay.Finally,according to the characteristics of cloud servers,we solve the computation delay of cloud computing layer.Experimental results show that the data processing delay of CFNAM is superior than that of the traditional cloud computing architecture model.(2)Optimization algorithm design of data processing energy consumption based on immune algorithm.Based on the traditional fog computing architecture,we propose a four-layer network architecture model,that is,a proxy fog server(PFoS)layer is added between cloud computing layer and fog computing layer.In this way,cloud data center can cache the data resource in the PFoSs in advance,and the PFoSs will provide the local service for fog nodes.We use the optimal immune algorithm to address the PFoSs,which not only can reduce the energy consumed by fog nodes in order to obtain data resources from the PFoSs,but also the number of fog nodes that the cloud server need to cache resources in advance.Experimental results show that compared with the traditional fog computing architecture,energy consumption is reduced by 44.95 % in the four-layer network architecture when the amount of data processing is 20 Gb.