Adaptive Multi-Threshold Energy-Aware Virtual Machine Consolidation in Cloud Data Center

2019 6th International Conference on Behavioral, Economic and Socio-Cultural Computing (BESC) Pub Date : 2019-10-01 DOI:10.1109/BESC48373.2019.8963569

Yingyue Hu, Ding Ding, Kaixuan Kang, Tingting Li

{"title":"Adaptive Multi-Threshold Energy-Aware Virtual Machine Consolidation in Cloud Data Center","authors":"Yingyue Hu, Ding Ding, Kaixuan Kang, Tingting Li","doi":"10.1109/BESC48373.2019.8963569","DOIUrl":null,"url":null,"abstract":"The ever-increasing energy consumption in cloud data centers not only translates to high operating costs, but also leads to negative impact on environment. Dynamic consolidation of virtual machine (VM) is proven to be an efficient way to improve resource utilization and reduce energy consumption in cloud data centers. In this paper, both the CPU utilization of system and SLA of users are taken into account to classify hosts and an adaptive multi-threshold energy-aware virtual machine consolidation algorithm is proposed to provide different consolidation mechanisms for different types of hosts. First, compound threshold is designed for overload hosts and will be adjusted dynamically to ensure both CPU utilization and SLA. Then a Q-Iearning based method is proposed to further divide underload hosts to save energy. Experiment results show that, our proposed algorithm can optimize resource utilization and reduce energy consumption of the data centers while minimizing the SLA violation rate and the number of migrations.","PeriodicalId":190867,"journal":{"name":"2019 6th International Conference on Behavioral, Economic and Socio-Cultural Computing (BESC)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2019 6th International Conference on Behavioral, Economic and Socio-Cultural Computing (BESC)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/BESC48373.2019.8963569","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

Abstract

The ever-increasing energy consumption in cloud data centers not only translates to high operating costs, but also leads to negative impact on environment. Dynamic consolidation of virtual machine (VM) is proven to be an efficient way to improve resource utilization and reduce energy consumption in cloud data centers. In this paper, both the CPU utilization of system and SLA of users are taken into account to classify hosts and an adaptive multi-threshold energy-aware virtual machine consolidation algorithm is proposed to provide different consolidation mechanisms for different types of hosts. First, compound threshold is designed for overload hosts and will be adjusted dynamically to ensure both CPU utilization and SLA. Then a Q-Iearning based method is proposed to further divide underload hosts to save energy. Experiment results show that, our proposed algorithm can optimize resource utilization and reduce energy consumption of the data centers while minimizing the SLA violation rate and the number of migrations.

查看原文本刊更多论文

云数据中心的自适应多阈值能量感知虚拟机整合

云数据中心的能源消耗不断增加，不仅导致运营成本上升，还会对环境造成负面影响。在云数据中心中，动态整合虚拟机是提高资源利用率和降低能耗的有效途径。本文结合系统CPU利用率和用户SLA对主机进行分类，提出了一种自适应的多阈值能量感知虚拟机整合算法，为不同类型的主机提供不同的整合机制。首先，复合阈值是为过载主机设计的，并会动态调整，以保证CPU利用率和SLA。然后提出了一种基于q学习的方法来进一步划分欠载主机以节省能量。实验结果表明，该算法在最小化SLA违规率和迁移次数的同时，优化了数据中心的资源利用率，降低了数据中心的能耗。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

2019 6th International Conference on Behavioral, Economic and Socio-Cultural Computing (BESC)

自引率

0.00%

发文量