云环境中虚拟机迁移策略的研究

发布时间：2018-11-21 20:13

【摘要】：保证用户的服务等级协议(SLA)与能源的高效利用是目前云计算极为关注的两大重点问题。虚拟化技术是云计算资源管理中的关键技术,其中,虚拟机迁移技术和策略都是极为引人关注的问题,本文将虚拟机迁移策略的优化作为研究的首要目标。针对如何有效降低大规模云数据中心的能量消耗并在一定程度上保证用户的服务等级协议(SLA)的问题,本文提出一种基于虚拟机迁移的优化算法LBES (Load Balancing and Energy Saving)。该算法综合考虑多种资源负载情况以及群聚冲突等问题,针对虚拟机迁移时机的确定,提出了基于预测模型的触发策略；针对待迁移物理机中存在的虚拟机群选择哪部分虚拟机进行迁移,提出了多资源约束的选择策略；并使用基于概率模型的定位策略解决了目标节点的选取问题。并在云模拟器中模拟的虚拟云数据中心对LEBS算法进行实现与分析。实验结果表明,在相同的阈值设置条件下,LEBS算法比其它算法在保证用户的服务等级和节约能耗方面的性能更优。
[Abstract]:To ensure the service level agreement (SLA) and the efficient use of energy are two important issues in cloud computing. Virtualization technology is the key technology in cloud computing resource management, among which, virtual machine migration technology and strategy are very interesting issues, this paper takes the optimization of virtual machine migration strategy as the primary goal of the research. Aiming at how to reduce the energy consumption of large scale cloud data center and ensure the service level protocol (SLA) of users to a certain extent, this paper proposes an optimization algorithm LBES (Load Balancing and Energy Saving). Based on virtual machine migration. The algorithm considers various resource loads and cluster conflicts, and proposes a trigger strategy based on prediction model to determine the migration opportunity of virtual machine. According to which part of virtual machines are selected for migration in physical machines to be migrated, a multi-resource constrained selection strategy is proposed, and the problem of selecting target nodes is solved by using the location strategy based on probability model. The LEBS algorithm is implemented and analyzed in the virtual cloud data center simulated in the cloud simulator. The experimental results show that the performance of LEBS algorithm is better than that of other algorithms in ensuring the service level of users and saving energy consumption under the same threshold setting condition.
【学位授予单位】：中南大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TP302

【参考文献】