Abstract: To maximize the economic income of grid system，a Bayes-based approach for grid resource allocation is presented，considering applications’ QoS requirements.The approach is based on the thinking that price can represent resource’s QoS comprehensive performance，by using Naive Bayes，allocates the resource of congruent charge range to the task according to the price parameter that the user submits，and gives priority to tasks of higher-level QoS，obligates resources of higher-level QoS currently to tasks of higher-level QoS later，ignoring tasks of lower-level QoS currently.The experiment shows that the proposed approach can not only meet users’ QoS requirements but also make higher economic income for grid system.

Key words: grid resource, resource allocation, Quality of Service（QoS）, income, Bayes

摘要： 考虑网格应用QoS需求，以最大化网格系统的经济收益为目标，提出了一种基于贝叶斯策略的网格资源分配方法。该方法基于价格可代表资源QoS综合性能的思想，利用历史QoS匹配记录，运用朴素贝叶斯定理根据用户提交的价格参数来分配与其级别相对应的符合要价范围资源，并优先考虑QoS水平较高的任务，将当前时刻QoS水平较高的资源预留给下一时刻到达的QoS水平较高的任务，而忽略当前时刻QoS水平较低的任务。实验结果表明，该资源分配方法不仅能有效地保障用户QoS，而且能使网格系统获得较大的经济收益。

关键词: 网格资源, 资源分配, 服务质量, 经济收益, 贝叶斯