Research on city Trust Evaluation Model Based on Grid Computing and Context Awareness

Abstract

The current evaluation mechanism of city confidence based on grid computing is not perfect, which will greatly affect the efficiency and the grid resources' sharing. On the base of the traditional city model of grid computing, in this paper we propose a context-aware control and city trust evaluation model, estimating through the context similarity, effectively estimating unfamiliar entities trust and providing incentive punishment mechanism based on the actual transactions. Resource sharing is the main features of grid. Grid resource management based on the economy computing promotes the sharing of grid resources from many perspectives. These management methods are based on certain precondition, which assumes the resources' transaction parties in grid market are reliable. This paper also proposed grid computing city architecture and evaluation process based on trust, reducing the impact from the false service of unfamiliar entity on the trust evaluation model. On this base people provide a variety of trust evaluation methods to achieve trust evaluation and management of entities. The main findings include trust evaluation based computing collaboration, context-aware inter-operable trust evaluation model, trust evaluation model grid services, trust-based evaluation of resource trading model. These trust evaluation model focuses on considering the validity of the trust, but lack of attention to the reliability of the grid entity trust evaluation when in the implementation. Due to the distribution of the grid system, heterogeneous and dynamic, the grid transactions often occur between unknown entities. Due to the strangeness of grid entities, the reliability of trust evaluation model in the actual run-time is reduced. To solve these problems, this paper presents a relatively reliable trust evaluation model in grid computing economy. When doing the trust evaluation on trustee, the trust-maker will enhance the accuracy of the evaluation trust between the strange entities by using trusted context and propose a mechanism for appropriate incentives and penalties trading entity thereby enhance the stability and reliability of trust evaluation model.