Energy and performance efficient resource allocation for cloud-based data centers

Cloud computing provides computing as a service form, due to this more andmore users migrated into the cloud instead of maintaining own physical infrastructure. It’s offering hardware, software, infrastructure as a service form to the users. Users have to pay as much as they use which also known as pay as you go model. Cloud computing facilitates sharing resources over the Internet. It is also a technology revolution, offering flexible IT usage in a cost efficient and pay-per-use manner. Cloud computing is a model for enabling convenient, on-demand network access to a shared pool of configurable computing resources such as, networks, servers, storage, applications,and services. These resources can be rapidly provisioned and released with minimal management effort. The underlying concept of cloud computing isthe separation of applications from the operating systems and the hardware on which they run. The Cloud computing proliferation has resulted in the establishment of large-scale data centers around the world containing thousands of computing nodes. But data centers consumed erroneous electrical energy for its huge hardware infrastructures which also responsible for carbon emission. This thesis demonstrates that the dynamic consolidations of virtualmachines in cloud data centers by proposing energy efficient algorithms and policies. The aim of this research is to propose a few new host overload detection and VMselection algorithms in order tominimize energy consumption. Recently,researchers have shown an increased interest to reduce energy consumption in cloud data centers by consolidation of Virtual Machines. The objective is to develop the quality of service constraints by the under workload for reducing energy consumption and to improve the application of computing resources. Dynamic VM consolidation leverages fine grained fluctuations in the application workloads, and continuously reallocates VMs using live migration to minimize the number of active physical nodes. Energy consumption is reduced by dynamically deactivating and reactivating physical nodes to meet the current resource demand. The proposed approach is distributed, scalable, and efficient in managing the energy and performance trade-off. This research clarifies the concerns about energy savingsi n cloud computing, analyzing the factors that make the reduction of energy consumption. Energy consumption, energy and service level agreement violation, service level agreement violation and number of virtual machine migrationwasmeasured to ensure service performancewhile energy consumption reduced. In this study, we found that our proposed algorithms able to reduce energy consumption by 3% and 19% for host overload detection and VM selection algorithm respectively