【摘要】 网格任务调度目标有很多,如用户要求任务轮转时间短、花费代价小,而资源提供者希望资源利用率高等,这些目标相互冲突,因此网格任务调度不仅是一个NP难问题,而且是一个多目标优化问题。本文根据网格环境下任务的时间相关性特点,对传统蚁群算法进行了改进,提出了一种快速网格任务调度算法。该算法不仅解决了网格调度中多目标优化问题,而且依据任务调度历史信息生成蚁群算法的初始信息素分布,提高了蚁群算法的求解速度。更多还原

【Abstract】 In a grid environment, the goals of the user and the resource provider can be conflicting, that is,the user may have the goals of minimizing turn-around time and the cost for the job, while a resource provider may have the goals of improving the throughput of the resources. So grid job scheduling is not only an NP-Hard problem, but also a multi-object optimization problem. In this paper, we propose a grid scheduler based on Ant Colony Algorithm, which satisfies the mutually conflicting goals of the users and the resource provider, as well as improves the velocity of the scheduler, by taking advantage of historical information on scheduling to form the initial information pheromone.更多还原