A Cloud Resource Scheduling Strategy Based On User Expectation And Cloud Platform Benefit

Cloud computing,with the features of flexible resource assignment,timely on-demand service and transparent by-quantity pricing,has been widely applied recently.Owing to the increasing concern about the cloud platform,the number of tasks submitted by users is also increasing and its pressure is soaring.Therefore,it is necessary to improve the processing efficiency and relieve the pressure of the platform.However,task scheduling and resource allocation scheme are the crucial factors to affect the efficiency of cloud computing.In fact,there have been many related studies on task scheduling and resource allocation in cloud computing.But most of them only take the benefit of the cloud platform provider into account.in other words,the scheduling scheme is designed by minimizing the maximum completion time,reducing energy loss and balancing load.In addition,on-demand service mode is used in cloud computing.Therefore,how to improve the quality of service is also an important study.At present,reducing the completion time of each task submitted by the user is often used to ensure the quality of service.Besides,as a business model with pay-as-you-go mechanism,there are more studies about pricing schema in cloud computing.In fact,a good resource scheduling scheme should be able to maximize the benefits while meeting the needs of the user.Therefore,users' demand can be quantified by considering the time and cost requirements of each task,and a resource scheduling model considering user needs and benefits of cloud platform providers is constructed to achieve a win-win goal.Abandoning the traditional pricing scheme which only relates to the length of tasks,a new dynamic pricing model which considers the actual needs of users and the characteristics of the cloud platform is constructed to realize the dynamic conversion between time and cost.Meanwhile,the concepts of user expectation and user satisfaction are put forward to quantify user needs,and user expectation acts as the constraints while user satisfaction is the optimization goal in the process of resource scheduling.In addition,due to users' different needs of the time and cost,two different weight allocation schemes when constructing the objective function are proposed.One only considers the high-demand side for time or cost and ignores the low-demand side while the other,which considers the time and cost demand of tasks,assigns the weight by the proportion of completion situation to the benchmark.A user expectation-oriented resource scheduling scheme that genetic algorithm is used to realize the resource scheduling process with constraints so as to improve user service quality is proposed.Finally,the simulation experiment is carried out to verify the validity of the dynamic pricing model,the reasonableness of the weight allocation scheme and the efficiency of the scheduling scheme.In view of the cloud platform provider,it is necessary to maximize the benefits of cloud platform by reducing the operation cost of cloud platform as lowly as possible.According to the analysis of the research status,the cloud benefit model is constructed by combining with the makespan,energy consumption and load.Moreover,minimizing the makespan can reduce the waste of resources,reducing energy loss can reduce cost and load balancing can ensure the stability of the platform.In addition,genetic algorithm is used to build a cloud benefit-orient resource scheduling model in order to provide a stable and smooth cloud computing service platform to better acquire user loyalty and earn greater benefits.Finally,the feasibility and effectiveness of the proposed algorithm are proved by contrast experiments.However,both of the two major players in cloud platform,whose interests are often conflicting,want to ensure their own interests.Therefore,a resource scheduling model based on user demands and cloud benefit is designed to measure the interests between users and cloud providers.The dynamic pricing model is constructed to guarantee the user quality of service.Meanwhile,the running cost is reduced to improve the cloud platform providers' benefit from the side.Besides,genetic algorithms is used to allocate resources to achieve the optimal solution,which does not only meet users' different needs as greatly as possible to improve the user satisfaction,but also considers the makespan,energy consumption,load balancing and other factors of task queue in cloud platform in order to ensure the stability of cloud platform and the fluency of running tasks.Finally,the simulation experiments have been done,and the results show that the proposed model has a sound effect.