Research On Cloud Fog Network And Its Load Balancing Strategies For Low Latency

As an important part of emerging information technology,the Internet of things(Io T)has become an important factor in the competition of comprehensive national strength among countries.The Internet of things is an interconnected network of things,which connect every objects to the Internet by various sensor equipment and systems,according to the contract agreement,to communicate and exchange information,in order to achieve intelligent identification,positioning,tracking,monitoring and management.Cloud computing as the support platform for data management and processing,promotes the development of the Io T technology.However,because cloud computing is often deployed in the center network which is far from the end users,and it is difficult to meet the demand for mobility support,geo-distribution,location awareness and low latency.Therefore,this paper is aimed at the shortcomings of cloud computing in supporting the application of Io T,and studies the fog computing technology which can make up the cloud computing.The characteristics and application of fog computing are analyzed emphatically.Besides,the cloud fog mixed network architecture,the architecture's latency modeling and the latency optimization modeling under energy consumption constraint are studied systematically.Furthermore,the corresponding latency optimal load balancing strategies are studied.The specific contents of this paper are as follows:A medical big data cloud fog mixed network architecture is proposed to solve the problem of high processing latency of the medical big data in cloud data centers.The architecture builds a fog computing layer by using edge devices such as routers or switches in a hospital.Fog computing devices proactively cache analysis results of medical images and other medical big data from cloud data centers,and analyze the data from medical terminals to get the diagnostic results to reduce processing latency.In this paper,the task processing latency of fog computing network is modeled.Meanwhile,modelling the task processing latency of fog network,and considering the weak computing power of fog devices,this paper proposes a latency-optimized load balancing strategy for fog network.The simulation results indicate that the fog network with constrained particle swarm optimization load balancing algorithm decreases the latency effectively.When using 10 fog devices and processing 100 medical image diagnostic requests,compared with a single fog node and the cloud computing network,the latency performance increases by 88% and 94% respectively.In order to solve the problem that cloud computing is difficult to meet the demand of more and more latency sensitive services,a network architecture of software defined cloud fog network in Internet of vehicles is proposed.The architecture integrates Software Defined Network(SDN),fog computing,cloud computing and Internet of vehicles(Io V),leveraging SDN to providing flexible centralized control and management,and using low latency,location awareness,mobility support etc.characteristics of fog computing to improve the quality of service in Io V.Moreover,this paper proposes the whole cloud data center as a distributed computing node to compose cloud fog distributed computing network and improve task processing efficiency.Moreover,the latency modeling of software defined cloud fog network and the latency modeling of fog network under energy consumption constraint of fog network are studied.Furthermore,two kinds of modified particle swarm optimization(MPSO)load balancing strategies are proposed to balance the load of cloud fog network and achieve the goal of optimization latency.The simulation results indicate that the software defined cloud fog network with MPSO2 strategy decreases the latency of Io V effectively.When using 10 fog devices,processing 5000 vehicle collision detection requests in Io V,compared with the fog network,a single fog node and the cloud computing,the latency performance of software defined cloud fog network increases by 23.3%,90% and 94% respectively.