Research And Evaluation Of Load Balancing Capability Of Cloud Platform

With the rapid development of cloud computing,cloud computing has been widely used in various fields of society.Many enterprises and organizations such as the Party,government and army have built cloud platforms for their own use or provide services to the outside world.Comparing with the traditional computing mode,cloud computing is cheaper and has high scalability.The various services introduced can greatly reduce the user's use threshold.However,cloud platform with cloud computing technology as its core is a multi-tenant and dynamic distributed system.Due to the improper application of resource scheduling strategy of cloud platform,cases of user experience decline occur from time to time.In addition,besides the cloud computing giants at home and abroad,the system stability,performance and resource utilization of other self-built cloud platforms are often not effectively guaranteed,which requires a reasonable evaluation method to help them self-evaluate and provide an optimization perspective.Load balancing is an important supporting technology and key service of cloud platform.A lot of work has been done in the academic field around load balancing algorithm,and it has been applied in cloud platform.However,there is little work to measure and evaluate the load balancing capability of cloud platforms.Especially when cloud platforms implement load balancing,they use many technologies synthetically,such as virtual machine scheduling and virtual machine migration,so they can't be evaluated from a single technical point of view.On the basis of detailed analysis of load balancing technology of cloud platforms,evaluation indexes and evaluation methods need to be put forward systematically.Thus,it can be used to test and evaluate the achievement degree of cloud platform system in real deployment and operation on the key index of load balancing.This paper focuses on the load balancing capability evaluation of cloud platforms.Firstly,this paper analyses the potential factors that may affect the load balancing capability of cloud platforms,and then proposes two different levels of cloud platforms,i.e.Iaa S layer and Saa S layer,to discuss the load balancing capability evaluation methods and evaluation system of cloud platforms.Secondly,by analyzing the different emphasis of Iaa S layer and Saa S layer on load balancing,the corresponding evaluation index,evaluation model,test load and evaluation process of each layer are designed.Finally,according to the proposed evaluation method and system,a prototype system of load balancing capability evaluation tool for cloud platform is designed and implemented,which provides users with visualization,process control,index monitoring and evaluation in one evaluation experience.This paper takes Open Stack cloud platform as an example to test and evaluate the load balancing capability of cloud platform.According to the evaluation system proposed in this paper,the evaluation results are obtained by analyzing and calculating the experimental data obtained after the experiment.After comparing the evaluation given by the model with the actual results,it is proved that the evaluation method and system proposed in this paper for load balancing capability of cloud platforms are reasonable and effective.