针对虚拟计算环境下资源交易的欺骗性和风险性,本文提出一种信任驱动的资源交易模型.首先基于Dempster-Shafer理论,给出一种信任度评估方法,然后对资源交易进行合理划分,在执行每步交易前评估双方信任度,依据其变化动态设定交易额,并确保每步资源的提供与支付同步进行,从而建立信任驱动的支持“边用边支付”策略的交易模型.仿真实验表明该模型能以一定的交易时间为代价,有效确保资源交易的公平性及安全性.
Virtual users face increasing cheat and risk when executing resource trades, which is taken as a prominent feature of the market-oriented virtual computing environment. So this paper proposes a trust-driven resource trade model to assist users to trade and avoid being cheated. First, a trust mechanism based on the Dempster-Shafer theory is built for evaluating trust degrees which reflect user behavioral characteristics. In this trade model, one whole trade can be divided into many steps, and before each of which, the trust degrees of both parities are evaluated, The trade granularity of each step depends on the fluctuation of users' trust degrees between the current step and the previous one. This model supports the "pay-as-you-use" strategy which allows resource consumers to pay resource providers step by step. Simulations show our proposed model can ensure users to execute resource trades more securely and trustworthily, with costing a few more time.