Multi-objective ACO resource consolidation in cloud computing environment

Cloud computing systems provide services to users based on a pay-as-you-go model. High volume of interest and a number of requests by user in cloud computing has resulted in the creation of data centers with large amounts of physical machines. These data centers consume huge amounts of electrical energy and air emissions. In order to improve datacenter efficiency, resource consolidation using virtualization technology is becoming important for the reduction of the environmental impact caused by the data centers (e.g. electricity usage and carbon dioxide). By using Virtualization technology multiple VM (logical slices that conceptually called VMs) instances can be initialised on a physical machine. As a result, the amounts of active hardware are reduced and the utilisations of physical resources are increased.;The present thesis focuses on problem of virtual machine placement and virtual machine consolidation in cloud computing environment. VM placement is a process of mapping virtual machines (Beloglazov and Buyya) to physical machines (PMs). VM consolidation reallocates and optimizes the mapping of VMs and PMs based on migration technique. The goal is to minimize energy consumption, resource wastage and energy communication cost between network elements within a data center under QoS constraints through VM placement and VM consolidation algorithms. The multi objective algorithms are proposed to control trade-off between energy, performance and quality of services. The algorithms have been analyzed with other approaches using Cloudsim tools. The results demonstrate that the proposed algorithms can seek and find solutions that exhibit balance between different objectives.;Our main contributions are the proposal of a multi-objective optimization placement approach in order to minimize the total energy consumption of a data center, resource wastage and energy communication cost. Another contribution is to propose a multiobjective consolidation approach in order to minimize the total energy consumption of a data center, minimize number of migrations, minimize number of PMs and reconfigure resources to satisfy the SLA. Also the results have been compared with other single-objective and multi objective algorithms.