Dynamic Voltage and Frequency Scaling-aware dynamicrnconsolidation of virtual machines for energy efficient cloudrndata centers

摘要

Computational demand in data centers is increasing because of the growing popularity of Cloudrnapplications. However, data centers are becoming unsustainable in terms of power consumptionrnand growing energy costs so Cloud providers have to face the major challenge of placing themrnon amore scalable curve. Also, Cloud services are provided under strict Service Level Agreementrnconditions, so trade-offs between energy and performance have to be taken into account. Techniquesrnas Dynamic Voltage and Frequency Scaling (DVFS) and consolidation are commonly usedrnto reduce the energy consumption in data centers, although they are applied independently andrntheir effects on Quality of Service are not always considered. Thus, understanding the relationshiprnbetween power, DVFS, consolidation, and performance is crucial to enable energy-efficientrnmanagement at the data center level. In this work, we propose a DVFS policy that reduces powerrnconsumption while preventing performance degradation, and aDVFS-aware consolidation policyrnthat optimizes consumption, considering the DVFS configuration that would be necessary whenrnmapping VirtualMachines tomaintainQuality of Service.We have performed an extensive evaluationrnon the CloudSim toolkit using real Cloud traces and an accurate powermodel based on datarngathered from real servers.Our results demonstrate that includingDVFS awareness in workloadrnmanagement provides substantial energy savings of up to 41.62% for scenarios under dynamicrnworkload conditions. These outcomes outperforms previous approaches, that do not considerrnintegrated use of DVFS and consolidation strategies.