The Research On Evaluation Model Of Cloud Service Based On QoS And Application

Cloud computing is a calculation mode, which provides storage resources, computing resources and software services to user by Internet and the users do not have to care about the various hardware and software resources. Cloud computing provides open standard, scalable system and service oriented architecture, which makes providers provide cloud service users reliable and on-demand service with a flexible and affordable way. The unique advantages of cloud service make more and more users use cloud service. With the developing of cloud service market, and there are more and more enterprises engaging in cloud computing R&D, service, network infrastructure and terminal equipment manufacturing. At the same time, the rapidly developing cloud technology promotes a variety of cloud services resource increasing ceaselessly. A large amount of cloud computing service providers have developed the same or similar functions cloud services. Thus, the system research on cloud service evaluation provides cloud service consumers basis to choose the appropriate cloud service product, which have the extremely important advantage.Based on the current cloud service evaluation study, this paper finds the current cloud service evaluation studies only establish cloud service evaluation model from the perspective of Web service or combination cloud service characteristics. The indexes in it are not comprehensive; parameter measurement is lacking, and the case studies only select SaaS service as object. Aiming at these problems, this paper establishes cloud service evaluation model based on QoS, which can effectively help the cloud service consumers choosing and making decision the same or similar functions cloud services product. In this paper, the main research content includes:Firstly, combining cloud service’s character that provider offer service based on the Internet service, this paper puts forward cloud service evaluation research angle based on non-functional service attributes. Reviewing the defects on cloud service evaluation research, the research focuses on evaluation model relying on Web service general evaluation model and combining cloud services feature. Then it elucidates the service evaluation foreground in cloud computing market.Secondly, from the non-functional service attributes, it inherits from existing Web service general evaluation index, references the existing cloud services evaluation index, combines with the characteristics of cloud services, adopts qualitative analysis and quantitative research methods, then establishes cloud service evaluation index system including response time, throughput, accuracy, robustness, mean time between failure, system stability, data management capabilities, service authentication, authorization and access, scalability and resource utilization in all11index.Then, QoS services evaluation problem is multiple attribute decision making problems. The indexes in established index system are not hierarchy structure but network structure characteristics. Based on this, it uses super decisions software constructing and solving cloud service evaluation model based on QoS by ANP calculation steps. It also calculates the weight of each index and comprehensive ranking.Lastly, this paper selects Amazon S3, ShengDa yun, Google Cloud Storage Aliyun OSS and Baidu cloud storage as the specific object to go on empirical research. The analysis result by combining the reality of cloud computing market validates the rationality of index system and model.