Data centres are the hubs for global connectivity through networking. Cloud computing has become an indispensable need to fulfil the insatiable everexpanding networking demand. More and more data centres are raised to fulfil this objective, consequently resulting in enhanced environmental pollution. Hence, server-related energy conservation is an indispensable need for minimizing carbon emissions. Factors like grid-related energy consumption and associated carbon emissions enhance electricity cost and carbon tax which results into total operating costs of data centres. The aim focuses on maximization of green energy usage with minimization of operating cost and carbon emission through process of virtual machine placement decision. Dynamic virtual machine consolidation emerges as a convincing solution for minimized energy consumption with optimized resource utilization in data centres. Virtual machine consolidation problems being truly NP-hard demand several heuristic algorithms for addressing the problem. Since most of the existing works are focussed for reducing number of hosts and their present resource utilization unmindful of future resource requirements, it results into unnecessary virtual machine migrations along with enhanced service-level agreement (SLA) violations. This problem can be duly addressed by considering current and future resource utilization by negotiating it as a bin-packing problem. The prediction of future resource utilization can be secured by using a k-nearest neighbour regression-based model.
Cloud computing inherits sharing of data from pool of resources existing in data centres when ever demanded. The imminent requirement for this purpose is proficiency of the data centre for fulfilment of this coveted objective. The pursuit of energy-efficient peak performance level is challenged by a simultaneous hike of energy consumption. The energy-efficient metrics contribute a major role for attainment of desired objective of safeguarding the environment. These metrics address the enhancement of the system’s proficiency. An increased energy-efficiency results into reduced consumption of energy resources since these energy resources are mostly non-renewable in nature and are the main source of carbon and heat emissions from operational data centres. As a matter of fact, any individual metric is not capable of achieving enhanced energy-efficient performance in a data centre. Therefore a collective utilization of selected metrics pertaining to power, performance and network traffic can improve the energy-efficient capability of data centre communication systems. The testing platform for such metrics is based on certain architectures which include D Cell, B Cube, Hyper Cube and Fat tree three-tier architectures.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.