Adaptive Threshold-Based Algorithm for Multi-objective VM Placement in Cloud Data Centers

Abstract

Data is grown exponentially day by day which increases demand in cloud storage, leads to setting up a large number of cloud data centers. Such data centers employ an immense amount of electrical energy has ended up in enormous operating cost. Dynamic virtual machine consolidation allows to optimize energy consumption and Service Level Agreement (SLA) Violations. The process of selecting which virtual machine should be located (i.e., executed) on each host of the data center is known as Virtual Machine Placement (VMP). The existing approaches focus on energy consumption, yet ignoring SLA violations at the time of Virtual Machine Placement. This paper addresses the problems in cloud data centers by proposing an Adaptive Four Threshold framework and KMI-MMT VM selection algorithm. The experimental results show that the proposed algorithm outperforms the existing algorithm in terms of maximum energy efficiency and minimum SLA Violation rate. It also reduces the number of VMs migrated while VM allocation.