Research On Trust Mechanism Based On Evidence Theory In Cloud Computing Environment

Trust issue is the fundamental of security interaction between nodes, mainly including problems of trust model and trust management. Trust mechanism in establishing trust relationships between entities is very fast and safe, as a kind of effective method to replace the traditional security mechanisms in a distributed system. With the popularization and development of Cloud computing, trust evaluation techniques are used in Cloud computing environment to study the node of resource allocation, communications safety, storage security and other aspects.At present, how to establish a trust model to adapt the cloud computing environment, how to help cloud computing service providers to establish an effective security mechanism to improve the quality of cloud computing services, and on the basis of obtaining the trust of the users to cloud computing service providers, this becomes the key to popularization of Cloud computing. Because of the complexity and dynamics of the Cloud computing environment, we can use the trust mechanism to assess the entity's trust problem in Cloud computing environment in order to improve the quality and reliability of the cloud computing service.In order to evaluate the performance of nodes in clouding computing environment to provide decision support for resource allocation and scheduling of cloud centers, this paper proposes a trust evaluation model based on Evidence Theory. In the process of trust evaluation, considering the large number of uncertain factors in cloud environment, trust relationship is divided into "trusted", "untrusted", "uncertain", where trust is evaluated from these aspects. On the basis of the entity's Central Processing Unit(CPU), memory, disk and network status, these information is fused to get the direct trust value. Then, according to the interaction between the entity and the cloud center, the indirect trust value is acquired. Finally, the comprehensive trust value is obtained by the weighted average of direct trust and indirect trust.This paper build a trust model based on evidence theory and carry on the simulation, each node status information is collected through Ganglia to obtain the direct trust, through interaction simulation to obtain the indirect trust, finally, through direct trust and indirect trust get the comprehensive trust. The simulation results show that the proposed trust model can accurately and effectively evaluate the trust.