Research On Trust Negotiation Mechanism Based On SLA In Cloud Environment

With the development of cloud computing technology,there are a variety of cloud services with different quality.Because it is lack of the subdivided standard of service in the industry,the user are likely to encounter service interruption,the service with low level,even the user data security issues in the process of using cloud services and some malicious user evaluations have had a serious impact on the credibility of cloud service.Therefore,it has become a hot issue to realize trusted negotiation between users and services in the field of service computing.The negotiation model based on SLA(Service Level Agreement)plays a more and more important role in the interaction.This paper focuses on how to realize the trust negotiation between the user and service entities,through the data standardization,the false information filtering,irresponsible evaluation filtering and other aspects to process the user's QoS evaluation information;In the process of the calculation of service reputation value,it is consider the distribution of weights,the user evaluation credibility,stability,and the analysis of the influencing factors.It will make the direct trust calculation more accurate;In the calculation of indirect trust value,it will through a trusted user recommend,find the trusted path and other ways to calculate accurate indirect trust value.It is realize the calculation of indirect trust value between user entities and service entities.This paper mainly innovates from the following aspects:1.In the calculation of service real credit value,this paper proposes a calculation method of indirect trust values oriented personalized service attribute.It can get the accurate direct trust value by analysis the interaction occurs time,the using frequency of service,the transaction amount and other factors.On this basis,the problem of indirect trust value calculation will be solved by the method of trusted tree.2.In the aspect of user privacy data protection,this paper proposes a dynamic interactive method based on SLA in cloud environment.Through judge the trust level of cloud service providers and the trust level of the users,it can refused to user access to services with high risk factor,reduce the accessing times of malicious user,prevent the excessive number of malicious evaluation,and avoid the users use the untrusted service again and again;It will determine the confidentiality level of the privacy data of cloud user according to the user's trust level.This method can control the evaluate behavior of users,make the rewards and punishments for cloud users,improve the service reputation value calculation accuracy and provide a credible environment for the interaction between users and cloud services.3.To realize the personalized selection of service,this paper analyzes the user evaluation weight by assigning different weights to the attributes of the QoS.It will reflect the user's different preference and affect the final service reputation of the calculated value directly.It will calculate the weight for several categories which the users interested,not only saves the computing resources,but also can achieve accurate results and reflect the user's personalized goals.4.In the aspect of stability service recommendation,this paper proposes a service recommendation method based on stability and user trust evaluation.It will solve the service selection problem for the users by the way of quantifying stability;filter the evaluations made by malicious user;reduce the impact which made by the malicious evaluation for the computing of service trust value.On the basis of indirect recommendation tree,this paper puts forward the concept of Top-k service recommendation network.It can calculates the recommendation value of service accurately and improve the accuracy of service reputation value calculation.