通过分析网格的使用模式,设计实现了网格入口软件WebGrid。WebGrid的研究与实现围绕网格监控和任务调度展开,采用了基于遗传算法的任务调度算法。该算法采用资源-任务的间接编码方式,通过DAG图获取子任务的层次关系,并将子任务按照层次深度排序,解决了种群的非法问题。在单一资源上采用短任务优先和父结点优先两个原则来安排子任务的执行次序,避免出现任务堵塞的现象。
The use mode of grid is defined in the dissertation, a grid portal - WebGrid, is designed and implemented. The research and implementation of WebGrid are based on grid monitoring and task scheduling, by research existing grid monitoring and task scheduling mechanism, combine with existing distribute technique, a task scheduling strategy based on GA (genetic algorithm) is developed. The chromosome - coding method and the operator of genetic algorithm are discussed in detail. The relationship between subtasks can be obtained through the DAG, and then the subtasks are ranked according to their depth- value, which can avoid the emergence of invalidate chromosomes. In the single resource, two principles are adopted to determine the sequence of the subtasks: one is the short - subtask prior; the other is parent - subtask prior. MyProxy mechanism is introduced in security managing module of WebGrid, the disagreement of Web security protocols and Globus's grid security infrastructure is resolved.