{"title":"Energy-performance optimisation for the dynamic consolidation of virtual machines in cloud computing","authors":"H. Li, T. Li, Z. Shuhua","doi":"10.1504/IJSOI.2018.10008967","DOIUrl":null,"url":null,"abstract":"Dynamic consolidation of virtual machines (VMs) can reduce energy consumption by switching idle hosts to sleep mode. However, to meet the quality of service of customers, it is necessary to achieve the trade-off between energy and performance. This paper first puts forward a new dynamic threshold adjustment method using the variation coefficient of historical CPU utilisation, actual CPU utilisation and million instructions per second requests by VMs in migration list. Furthermore, it devises a novel VM allocation policy based on the grey correlation degree model, and formulates a conversion model of CPU utilisation for achieving better trade-off between energy consumption and performance. Finally, some experiments are carried out on the CloudSim and the PlanetLab workloads. The experimental results show that the methods proposed in this paper have obvious advantages on the trade-off between energy and performance during the VM consolidation.","PeriodicalId":35046,"journal":{"name":"International Journal of Services Operations and Informatics","volume":"9 1","pages":"62"},"PeriodicalIF":0.0000,"publicationDate":"2018-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Services Operations and Informatics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1504/IJSOI.2018.10008967","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"Business, Management and Accounting","Score":null,"Total":0}
引用次数: 3
Abstract
Dynamic consolidation of virtual machines (VMs) can reduce energy consumption by switching idle hosts to sleep mode. However, to meet the quality of service of customers, it is necessary to achieve the trade-off between energy and performance. This paper first puts forward a new dynamic threshold adjustment method using the variation coefficient of historical CPU utilisation, actual CPU utilisation and million instructions per second requests by VMs in migration list. Furthermore, it devises a novel VM allocation policy based on the grey correlation degree model, and formulates a conversion model of CPU utilisation for achieving better trade-off between energy consumption and performance. Finally, some experiments are carried out on the CloudSim and the PlanetLab workloads. The experimental results show that the methods proposed in this paper have obvious advantages on the trade-off between energy and performance during the VM consolidation.
期刊介绍:
The advances in distributed computing and networks make it possible to link people, heterogeneous service providers and physically isolated services efficiently and cost-effectively. As the economic dynamics and the complexity of service operations continue to increase, it becomes a critical challenge to leverage information technology in achieving world-class quality and productivity in the production and delivery of physical goods and services. The IJSOI, a fully refereed journal, provides the primary forum for both academic and industry researchers and practitioners to propose and foster discussion on state-of-the-art research and development in the areas of service operations and the role of informatics towards improving their efficiency and competitiveness.