Research And Implementation Of Hybrid Cloud Service Platform For QoS

As a new business model,cloud computing has won the favor of many consumers because of its high quality and low price and ready-to-use.However,in the case of more and more cloud service providers,users hope to have a platform for unified management of cloud services.And still get good service.For cloud service providers,cloud service providers hope to sell as many services or resources as possible to increase revenue.Users need high-quality and low-cost services and cloud service providers hope to sell more resources,which is a contradiction in terms of user service quality.In this contradiction,this thesis combines multiple service providers to ensure that the user's service quality does not exceed the expected worst quality of service.Under this premise,the user is requested to allocate a service provider.When the user request is allocated,not onley the acceptable worst quality of service as a constraint,but also how to reduce the idle resources of the service provider,so in order to ensure the quality of service of users,firstly,this thesis introduces the queuing theory model and proposes a two-tier queuing model.The first layer queuing model of the model can evaluate the user's request,and decide whether to accept the user request according to the evaluation result.In order to shorten the average waiting time of the user request,multiple queuing models are combined to form the second layer of the model.The queuing model shows that the model can reduce the waiting time and response time of the user request compared with the general queuing theory model.Secondly,in order to balance the user's service quality and the service provider's benefit,This thesis introduces the genetic algorithm and tabu search algorithm in multi-objective optimization algorithm,and proposed a hybrid algorithm of genetic algorithm and tabu search algorithm,namely GT(Genetic algorithm and Tabu search algorithm)algorithm,shows that the algorithm is better than the genetic algorithm and tabu search algorithm in convergence speed and solution;finally,this thesis develops a Hybrid cloud service platform,which uses the above-mentioned models and algorithms to allow access control of requests before user requests arrive,and then analyzes the services required for the requests.The platform can balance the quality of service of the user and the interests of the service provider when allocating the service provider to the service required by the user.The test results show that the hybrid cloud service platform can ensure that the user's service quality does not exceed the expected worst service quality,and at the same time,it can also take into account the interests of the service provider,reduce the idle resources of the service provider,and increase the service provider's revenue.