论文部分内容阅读
大多分布式资源调度策略的目标或为提高分布式系统的运行效率,或为提高用户的满意度.文中提出了一种综合考虑资源可信度与用户满意度的调度策略,以用户的满意度为目标函数,且将资源可信度作为用户满意度组成的一部分,通过遗传算法实现优化调度.继而将此调度策略封装进Agent之中,并提出了一种基于Multi-agent分布式系统架构,该架构由用户、作业和资源3类Agent组成,并且具有良好的分布式群集智能效应和自适应能力.算例结果显示,此调度策略不仅提高了系统运行效率,同时增加了用户满意度.
Most of the distributed resource scheduling strategy aims to improve the operating efficiency of distributed systems or to improve user satisfaction.In this paper, we propose a scheduling strategy that takes into account the resource trustworthiness and user satisfaction, with user satisfaction As the objective function, and regards the resource credibility as a part of the user satisfaction degree to achieve the optimal scheduling by genetic algorithm.Secondly, the scheduling strategy is encapsulated into Agent, and a multi-agent distributed system architecture is proposed, The architecture consists of three types of agents: users, jobs and resources, and has good distributed cluster intelligence and adaptability.The results show that this scheduling strategy not only improves the efficiency of the system, but also increases user satisfaction.