AbstractsPhysics

Studies on energy efficient techniques for cloud data centers;

by Anandharajan T R V




Institution: Anna University
Department: Studies on energy efficient techniques for cloud data centers
Year: 2015
Keywords: Energy Curve Model; Energy per Instruction Rate; Minimum Processing Power; Multi Informative VM Analysis
Record ID: 1196910
Full text PDF: http://shodhganga.inflibnet.ac.in/handle/10603/34201


Abstract

newlineThe objective of the research is to draw from existing approaches and techniques new insights that can assist the problem of power and performance trade off that exists in cloud data centers These insights would enable us to design our novel self adapting mechanism to solve the trade off We design a framework for the cloud environment and handle workload of virtualized servers and analyse the energy perspective A study was carried out on the available literature on data centers and the energy perspective for them That is the self managing mechanism should cater for the wide variation in reliability attributes and associated constraints for a large number of applications that are composed of services offered by a cloud environment Problem of dynamic virtualized instances selection have limitations when they need to scale to the case of the cloud Results of the literature review identified the axes that self adaptive VM consolidation need to be considered thus achieving better energy efficiency newlineThis thesis views the IaaS in the cloud based services and we propose a novel self adaptive energy efficient mechanisms to solve the trade off A Multi Informative VM Analysis MIVA was proposed to analyze the virtualized cloud component complexity newline%%%reference p162-168.