An adaptive heuristic for managing energy consumption and overloaded hosts in a cloud data center

摘要: In this paper, we address the problems of massive amount energy consumption and service level agreements (SLAs) violation in cloud environment. Although most existing work proposed solutions regarding SLA for data centers (CDCs), while ignoring some important factor: (1) analysing robustness upper CPU utilization threshold which maximize resources; (2) prediction based VM selection from overloaded host reduce performance degradation time violation. context, adaptive heuristic algorithms, namely least medial square regression detection minimum hosts. These algorithms reducing CDC with minimal SLA. Unlike algorithm consider types application running it at different periods over VMs. The approaches are validated using CloudSim simulator through simulations days a real workload trace PlanetLab.