论文部分内容阅读
鉴于复杂云环境具有较强的动态变化与分散性特点且云数据中心各实体又具有离散特性,提出了一种基于分布式离散事件系统监控理论的云资源动态调度策略.先将云环境构建为一个由物理服务器类、虚拟机类、云任务类等多个子系统构成的分布式离散事件系统(DES);再将分布式离散事件系统监控理论应用于云资源动态调度.通过系统局部控制器的局部控制及其与全局控制器之间的信息交互,实现对云资源调度的整体优化,保障各个子系统资源分配的合理性.同时,系统通过调控资源负载均衡,提高云资源整体利用率与云服务质量,提升整个云平台性能.实验结果表明,在任务量不断增加的情况下,所提的云资源动态调度策略与传统云资源调度策略相比,在资源利用率和任务执行时间等方面都具有较明显的优势.
In view of the strong dynamic change and decentralization of complex cloud environment and the discrete nature of cloud data center entities, this paper proposes a dynamic scheduling strategy based on distributed discrete event system monitoring theory.Firstly, the cloud environment is constructed as A distributed discrete event system (DES) composed of a number of subsystems, such as physical server class, virtual machine class and cloud task class, and then applies the distributed discrete event system monitoring theory to the dynamic scheduling of cloud resources. Through the system local controller Local control and information interaction with the global controller to achieve the overall optimization of cloud resource scheduling to ensure the rationality of resource allocation in each subsystem.At the same time, the system improves the overall utilization of cloud resources and cloud by regulating resource load balancing The quality of service and the performance of the whole cloud platform.The experimental results show that compared with the traditional cloud resource scheduling strategy, the proposed dynamic scheduling of cloud resources in terms of resource utilization and task execution time Has obvious advantages.