Energy-efficient resource management of cloud datacenters under fault tolerance constraints

摘要

The cloud computing as a ubiquitous paradigm could provide different services for internet users and Information Technology (IT) companies through datacenters that located around the world. However, cloud provider faces several problems such as high energy consumption and fault occurrence issues in cloud datacenters. Hence, cloud provider has to make a trade-off between energy and fault to gain more profit. In this paper, a migration method for the virtual machines to handle fault problem and prevent Service Level Agreement (SLA) violation is proposed by considering energy consumption constraints. The approach considers the lowest increasing in energy consumption and the minimal deadline miss ratio as the most significant factors for migration of each Virtual Machine (VM). The results show that low SLA violation is achieved with the least amount of energy consumption compared to other methods. It is also shown that the energy increasing by migration has an exponential relationship with the failure rate increasing.