Energy-Efficient Management of Data Center Resources for Cloud\n Computing: A Vision, Architectural Elements, and Open Challenges

Abstract

Cloud computing is offering utility-oriented IT services to users worldwide.\nBased on a pay-as-you-go model, it enables hosting of pervasive applications\nfrom consumer, scientific, and business domains. However, data centers hosting\nCloud applications consume huge amounts of energy, contributing to high\noperational costs and carbon footprints to the environment. Therefore, we need\nGreen Cloud computing solutions that can not only save energy for the\nenvironment but also reduce operational costs. This paper presents vision,\nchallenges, and architectural elements for energy-efficient management of Cloud\ncomputing environments. We focus on the development of dynamic resource\nprovisioning and allocation algorithms that consider the synergy between\nvarious data center infrastructures (i.e., the hardware, power units, cooling\nand software), and holistically work to boost data center energy efficiency and\nperformance. In particular, this paper proposes (a) architectural principles\nfor energy-efficient management of Clouds; (b) energy-efficient resource\nallocation policies and scheduling algorithms considering quality-of-service\nexpectations, and devices power usage characteristics; and (c) a novel software\ntechnology for energy-efficient management of Clouds. We have validated our\napproach by conducting a set of rigorous performance evaluation study using the\nCloudSim toolkit. The results demonstrate that Cloud computing model has\nimmense potential as it offers significant performance gains as regards to\nresponse time and cost saving under dynamic workload scenarios.\n