Research On Service Chain Deployment Mechanism Based On Cloud-Fog Collaboration In Smart Integration Identifier Networking

With the rapid development of the Internet,because of its "static" and "rigid" design defects,the traditional network architecture has been difficult to meet the increasingly diverse service requirements.Smart Integration Identifier Networking(SINET)proposes the design concept of hierarchical and domain division and intelligent integration which provides a new solution for solving traditional network architecture problems.In addition,network function virtualization(NFV)can provide the ability to decouple software and hardware,making the deployment of service function chaining(SFC)flexible and fast.Therefore,SINET uses NFV technology to build SFC to build functional family to carry services.In the traditional method,the SFC is mostly deployed in the cloud computing environment.However,with the increasing number of devices with heterogeneous resources and diversified service requests in the SINET,the traditional method cannot make full use of the heterogeneous resources in the network.And it is difficult to meet delay-sensitive service requests.Therefore,it is necessary to introduce fog computing as a supplement to cloud computing to provide available resources with low latency characteristics for services.This dissertation proposes the deployment of service chains based on the cloud-fog collaboration in the SINET,aiming to reduce the end-to-end delay of services,improving resource utilization,and meeting various service needs of users.This dissertation has designed and implement a service chain deployment mechanism based on cloud-fog collaboration in SINET.The mechanism combines the cloud-fog collaborative architecture to intelligently schedule and dynamically deploy the service chain,and proposes two heuristic algorithms for delay-sensitive and delaytolerant services.The main contents of this dissertation are as follows:(1)This dissertation analyzes the requirements of the service chain deployment mechanism based on cloud-fog collaboration,determines the node cost and service delay as the optimization goals,and designs a SFC deployment mechanism that is aware of service requirements and cloud-fog resources,including processing center module,services and Information monitoring module,information storage module,deployment decision module and SFC mapping module.(2)This dissertation has implemented and analyzed the service chain deployment mechanism based on cloud-fog collaboration.Through the mutual cooperation of various modules,the service chain can be optimized and deployed for different types of service requirements.User requirements are divided into delay-sensitive type and delaytolerant type,and proposes two different deployment algorithms,which jointly use cloud and fog resources to deploy service functions on demand in cloud nodes and fog nodes,which solves the problem of uneven resource utilization,which leads to excessive service delay or node load heavy problem.(3)This dissertation conducts functional tests and performance tests on each module of the deployment mechanism in the SINET prototype system and simulation environment.The functional test results show that the mechanism can provide deployment strategies according to the needs of users who are sensitive to latency or tolerant to latency.Compared with the deployment method in the cloud or fog computing environment,the deployment mechanism based on cloud-fog collaboration has better performance in indicators such as end-to-end delay,node cost and deployment success rate.This result also proves the effectiveness of this mechanism under the cloud-fog collaboration.