摘要： 针对传统网格任务调度模型将每个任务的预期执行时间设定为一个确定常量，无法表达其不确定性的问题，利用一种新的软计算方法——集对分析联系数来表示并处理网格任务预期执行时间的综合不确定性。提出基于联系数的不确定性网格动态调度模型，在线动态调度算法OUD_MCT和批模式动态调度算法BUD_Surferage。数值仿真结果表明，该类算法能较好地描述网格任务预期执行时间的动态性和不确定性，使传统网格调度方法成为其特例，在动态和不确定网格环境中具有良好的理论意义和实用价值。

关键词: 计算网格, 不确定性, 动态, 任务调度, 调度算法

Abstract: Aiming at the problem of traditional grid task scheduling model can not express and process the uncertainty existing in grid environment, because it takes the expected time to compute of a task as an constant, and by means of connective number of set pair analysis, this paper presents a new soft computation method to express and process the synthetic uncertainty of task scheduling for computing grid, namely using a connective number to express the expected time to compute of a task. After presenting the uncertain scheduling model based on connective-number, online uncertain dynamic scheduling algorithm OUD_MCT, and batch uncertain dynamic scheduling algorithm BUD_Surferage are presented for the uncertain dynamic computing grid. Numerical value simulation results illustrate that these algorithms can express the dynamics and uncertainty of expected time to compute of tasks in the computing grid environment. These algorithms are the generalization of traditional grid scheduling algorithms, and with high value in theory and application in the uncertain and dynamic grid environment.

Key words: computing grid, uncertainty, dynamic, task scheduling, scheduling algorithm

中图分类号: 

• TP393.07