Design Of Diverse Policies Supported Intent-Based Networking

With the research on Software defined networking and Network Function Virtualization going deep and the urge of effective and convenient network arrangement becoming stronger,study in Intent-Based Networking(IBN)is becoming a research hotspot of the network control and management area.IBN technology is a new approach to manage the network.Using this technology,network users can just tell the network the requirement to be meet,namely the intent,regardless of the technical details and IBN can automate manage,configurate the network and provide services.However,there are still some problems to be solved such as how to properly configure the intents as many as possible in the network when physical resources are limited,and for these unconfigured intents due to lack of network resources,how to properly reduce the resource demand of these intents to be configured again.The author emphatic researched the configuration of the time-varying intents appeared in the Intent-Based Networking and the negotiation strategy of these left out intents by a shortage of resources.The main work and contributions in this thesis are stated as below:(1)This thesis focussed on the configuration of the intents in the Intent-Based Networking,and analyzed the characteristic of the temperal intents in the internet that supports diverse policies.Then the author explained the optimization problem and proposed the heuristic algorithm based on the configuration algorithm that has existed to achieve the optimization goal.The goal of the algorithm is to maximize the number of intents configurations.The proposed algorithm is a greedy approach that consisted of TimePeriod Select Algorithm and Complie in single TimePeriod Algorithm,and the TimePeriod Select Algorithm is used to find the time period which the number of intents is largest,then the Complie in single TimePeriod Algorithm can configure the intents in any time period.The emulation on differ-scale networks demonstrates that the proposed intent configuration algorithm is more effective than the existing one in achieving the optimization goal.Also,the running time of the proposed algorithm is less than the existing one.(2)The thesis focussed on the negotiation of the intents in the Intent-Based Networking and compared the existing negotiation approaches to the temporal intents.The author extended the dimension of temporal intents from bandwidth to time and bandwidth,and then established optimize model.Based on this optimize model,the author proposed two different ways to the negotiation problem of the unconfigured temporal intents.One is Time-Bandwidth Negotiation Algorithm,this algorithm is consist of Time Period Change Algorithm and Bandwidth Negotiation for Single Intent Algorithm,the Time Period Change Algorithm is used to choose the proper time for the intent and the Bandwidth Negotiation for Single Intent Algorithm is used for calculating the proper bandwidth decrease of the intent.The other algorithm is to resolve the Optimize model with heuristic approach,the first step of algorithm is to find the possible time change values for all intents that need to be reconfigured and then calculate the time change rates and the bandwidth decrease rates for all the unconfigured intents.The author then realized the algorithms via computer program and compared the two algorithm through the emulations.The result shows,when the bandwidth decrease rate is more important than the time change rate,the algorithms proposed works better than just negotiation in bandwidth.