基于开源平台的云资源管理系统的研究与应用

发布时间：2017-12-30 19:39

本文关键词：基于开源平台的云资源管理系统的研究与应用　出处：《电子科技大学》2014年硕士论文　论文类型：学位论文

更多相关文章： 云计算 资源调配 负载均衡 虚拟机

【摘要】：随着云计算的发展与普及,用户对海量计算资源的需求与日俱增,而对于提供云计算资源的企业而言,如何对云计算平台中的资源进行稳定高效的管理成为越来越重要的问题。另一方面开源云平台软件虽然因免费开放的特点,受到了越来越多企业的关注,但其资源管理功能仍不完善,其管理的自动化程度以及机制的合理性、高效性仍有待完善。为了解决以上问题,本文以某企业云平台系统课题为背景,研究在Open Stack和Cloud Foundry开源平台中,如何实现有效的云资源管理,并开发实现云计算资源管理系统。资源管理系统首先实现对虚拟机、服务器资源负载的实时监控,记录存储监控数据,并以此为基础,实现平台资源自动化的弹性伸缩管理功能。通过分析虚拟机资源使用的监控记录,对虚拟机一段时间内的资源使用量进行预测,制定资源调配策略,恰当及时的扩展或收缩资源,保障虚拟机以及其中应用的运行稳定以及平台资源得到高效利用。同时,针对开源云平台中负责运行应用程序的虚拟机,即运行节点虚拟机,其负载均衡性的不完善,管理系统对虚拟机任务,即应用实例的分发机制进行优化。通过考虑每种应用实例资源消耗的特性,动态的制定实例分发策略,并且修正网络延迟对实例分发决策制定产生的误差,改善运行节点虚拟机的负载均衡状况。最后资源管理系统结合服务器与虚拟机资源的监控信息,综合考虑虚拟机与服务器各类型资源的负载情况,根据系统设定的策略,调用开源平台提供的接口,实现自动化的虚拟机热迁移功能,提升服务器的负载均衡性。本论文围绕为云计算平台提供稳定高效的资源管理的目标,完成开源云平台的搭建与资源管理系统的开发,提供较为自动化的资源管理功能,同时,提出了一种应用实例分发的优化机制,改善运行节点虚拟机的负载均衡状况,另外通过对虚拟机迁移策略的研究,综合分析虚拟机和服务器的综合运行状况,实现自动化的虚拟机热迁移功能。系统测试实验表明,本平台研制达到设计要求。
[Abstract]:With the development and popularization of cloud computing, users' demand for massive computing resources is increasing, and for enterprises providing cloud computing resources. How to manage resources stably and efficiently in cloud computing platform has become more and more important. On the other hand, open source cloud platform software has attracted more and more attention because of its free and open characteristics. But its resource management function is still not perfect, its management automation degree and mechanism rationality, high efficiency still need to be improved. In order to solve the above problems, this paper takes the cloud platform system as the background. This paper studies how to realize the effective cloud resource management in Open Stack and Cloud Foundry open source platform. The resource management system firstly realizes the real-time monitoring of virtual machine and server resource load, records and stores the monitoring data, and based on this, develops and implements the cloud computing resource management system. Through analyzing the monitoring records of the use of virtual machine resources, the resource usage of the virtual machine for a period of time is predicted, and the resource allocation strategy is formulated. Appropriate and timely expansion or contraction of resources to ensure the stability of the virtual machine and its applications, as well as the efficient use of platform resources. At the same time, the open source cloud platform is responsible for running the application of virtual machines. That is, running node virtual machine, its load balance is not perfect, the management system for virtual machine tasks, that is, the distribution mechanism of application cases is optimized. By considering the characteristics of resource consumption of each application instance. Make case distribution strategy dynamically, and correct the error caused by network delay to instance distribution decision making. Finally, the resource management system combines the monitoring information of server and virtual machine resources, and considers the load of virtual machine and server resources comprehensively. According to the strategy set by the system, the interface provided by the open source platform is called to realize the automatic heat transfer function of virtual machine. This paper focuses on the goal of providing stable and efficient resource management for cloud computing platform and completes the construction of open source cloud platform and the development of resource management system. At the same time, an optimization mechanism of application instance distribution is proposed to improve the load balance of virtual machine in running node. In addition, the migration strategy of virtual machine is studied. The integrated running condition of virtual machine and server is analyzed synthetically to realize the heat transfer function of automatic virtual machine. The system test results show that the platform meets the design requirements.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP393.09;TP302

【相似文献】