The Research On Vague Set-Based Grid Trust Model And Task Scheduling Algorithm

The concept of grid computing is becoming a more and more important in the high performance computing world. It enables the sharing, selection, and aggregation of geographically distributed heterogeneous resources for solving large-scale problems in science, engineering, and commerce. However, sharing is a volunteer action, and peers are not responsible for their irresponsible bartering history, therefore it is difficult to establish the trust relationship among users with traditional trust mechanism. Because of the sources in the grid are heterogeneous and geographically distributed, the management of task scheduling in such a large-scale distributed environment is a complex problem. Therefore, trust model and the performance of task scheduling algorithm will affect the efficiency and throughput of the grid system directly.We propose a new trust model based on Vague set theory and combine the trust model with task scheduling algorithm of relevant tasks. The contributions of this paper are as follows:1. Referring to social people trust relationship models and Bayesian networks-based trust model, we propose a subjective trust model based on Vague set theory. With Vague set theory and Bayesian networks-based theory we describe a method to establish trust ,which is based on subjective evaluation of the quality of services.2. The uncertainty of Gird users, recourses and services may play a negative affect on the execution of Grid tasks. In order to reduce this impact, we propose trust-update mechanism and punishment mechanism. Mathematic analyses and simulations show that, compared to the current model, the proposed model is more robust on trust security.3. Combine the trust model with task scheduling algorithm of relevant tasks. Adjust the structure of the algorithms, through add the confidence factor to the algorithms. Simulations on computing-intensive and data-intensive applications prove that the algorithms can efficiently satisfy the QoS requirement in trust, with costing a few more time.