The dynamic virtual machine (VM) consolidation is a key resource management technique used for achieving a trade-off between performance and energy efficiency of cloud computing systems. Selection of the source and destination hosts is a key problem in the process of VM consolidation which faces different challenges in the effective management of cloud resources. One of these challenges is utilizing burstiness-aware algorithms for selecting the source and destination hosts to prevent frequent migrations, reduce Service Level Agreement (SLA) violations, and improve energy efficiency. In the current study, we propose a new approach based on the Kullback-Leibler Distance (KLD) criterion to tackle this problem. The proposed approach includes burstiness-aware algorithms for selecting the source and destination hosts to minimize SLA violations, improve energy efficiency, and decrease the number of migrations. We utilize both real-world and random workloads and CloudSim simulator to inspect the performance of proposed algorithms. The experimental results reveal that the proposed approach outperforms the previous algorithms in terms of performance.

动态虚拟机（VM）整合是一种关键的资源管理技术，用于实现云计算系统的性能和能效之间的折衷。源主机和目标主机的选择是VM整合过程中的关键问题，在有效管理云资源方面面临着不同的挑战。这些挑战之一是利用突发性感知算法来选择源主机和目标主机，以防止频繁迁移，减少违反服务水平协议（SLA）的行为并提高能源效率。在当前的研究中，我们提出了一种基于Kullback-Leibler距离（KLD）准则的新方法来解决此问题。提议的方法包括突发性感知算法，用于选择源主机和目标主机，以最大程度地减少违反SLA的行为，提高能源效率，并减少迁移数量。我们利用现实和随机工作负载以及CloudSim模拟器来检查所提出算法的性能。实验结果表明，该方法在性能上优于以前的算法。