Abstract:

Aiming at the trust problem existing in cloud computing environment, we first propose a subjective trust model based on the Bayesian method to quantify and evaluate the trustworthiness of computing nodes, and demonstrate its mathematical description and implementation. Duo to the characteristics of dynamic, heterogeneity and deception, resource nodes are inevitably unreliable in cloud environments. So we also introduce a punishment mechanism and a pruningfiltering mechanism. We finally propose a dynamic level scheduling algorithm based on a Bayesian subjective trust model named BSTDLS by integrating the existing DLS algorithm. Theoretical analyses and simulation experimental results prove that the BSTDLS algorithm can efficiently improve the ratio of successful execution at the cost of sacrificing fewer schedule length.

Key words: cloud computing;bayesian estimation;trustworthiness;recommendation trust