Migration-Based Elastic Consolidation Scheduling in Cloud Data Center

Virtualization and cloud computing technologies now make it possible to consolidate multiple online services, which are packed in Virtual Machines (VMs), into a smaller number of physical servers. However, it is still a challenging scheduling problem for cloud provider to dynamically manage the VM allocation for handling variable workloads without Service Level Agreement (SLA) violation. In this paper, we propose a Migration-based Elastic Consolidation Scheduling (MECS) mechanism to automate elastic resource scaling for cloud systems. Different from the previous researches, we take both the dynamic workload fluctuation and the VM migration overhead into account. We first develop an online resource demand predictor, which is an ARIMA-based VM resource demand state predictor, to achieve adaptive resource allocation for cloud applications. We then propose a migration-based elastic consolidation scheduling heuristic to dynamically consolidate the VMs with adaptive resource allocation for reducing the number of physical machines. Extensive experiment results show that our scheduling is able to realize elastic resource allocation with acceptable effect on SLAs.