4.6 Article

Virtual Machine Migration Techniques for Optimizing Energy Consumption in Cloud Data Centers

期刊

IEEE ACCESS
卷 11, 期 -, 页码 86739-86753

出版社

IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/ACCESS.2023.3305268

关键词

Energy consumption optimization; virtual machine migration techniques; dynamic threshold; virtual machine selection; host selection; cloud data center

向作者/读者索取更多资源

The energy used by cloud data centers (CDCs) is increasing as cloud services expand, posing environmental burdens and higher expenses for cloud providers. To reduce energy consumption, virtualization migration and consolidation techniques are widely used in CDCs. This study proposes algorithms for determining migration timing, selecting VMs to migrate, and choosing migration destination hosts. Experimental results show that the proposed algorithms outperform state-of-the-art methods, reducing energy consumption, service level agreement violation, and the number of VM migrations.
The energy used by cloud data centers (CDCs) to support large volumes of data storage and computation is dramatically increasing as the scope of cloud services continues to expand. This puts a greater burden on the environment and results in higher expenses for cloud providers. Virtualization migration and consolidation have been widely used in current CDCs to achieve service consolidation and reduce energy consumption (EC). This study divides the fundamental tasks of virtual machine (VM) migration into three portions: determining migration timing, choosing the VMs to migrate out, and selecting the migration destination hosts. An EC levels-based adaptive dynamic threshold method for determining migration timing was proposed, as well as a correlation and utilization-based strategy for selecting the VMs to migrate out and an improved EC-aware best-fit algorithm for selecting the migration destination hosts. The pro-posed algorithms were evaluated using the CloudSim toolbox, and the real VM workload traces from PlanetLab were used as experimental data. According to the experiments, the proposed algorithms reduce EC, service level agreement violation (SLAV), and the number of VM migrations by an average of 15.49%, 7.85%, and 83.32% in comparison to the related state-of-the-art methods and benchmark algorithms. This suggests that the proposed methods outperform other techniques for VM migration, even when the workload necessitates a significant number of VMs or a greater amount of host resources, and improve the quality of service while optimizing energy consumption. However, the experiments were conducted in a simulation platform, which has some drawbacks, leading to the experimental results varying slightly from the actual environment.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.6
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据