合并(业务)
计算机科学
服务质量
业务
操作系统
电信
会计
标识
DOI:10.1504/ijbic.2024.141456
摘要
Traditional energy consumption optimisation strategies that do not consider the quality of service (QoS) may lead to customer loss and large profit loss. To solve the above problems, we propose a QoS-and-global-energy aware virtual machine (VM) consolidation strategy (QGE-VMC) for data centres. First, during the overload detection phase, we dynamically set overload thresholds for servers based on the historical utilisation of VMs to prevent host overload, which effectively reduces service level agreement (SLA) violations and improves QoS. Second, in the VM selection phase, we considered reducing both the time and number of VM migration, which reduces the time when servers are overloaded. Third, during the target host selection phase, we use a heuristic algorithm to make the selected servers ensure that the global energy consumption of the data centre rises to the minimum and will not be overloaded, thereby ensuring QoS while reducing data centre energy consumption.
科研通智能强力驱动
Strongly Powered by AbleSci AI