Home // International Journal On Advances in Internet Technology, volume 8, numbers 1 and 2, 2015 // View article
Authors:
Kai Spindler
Sven Reißmann
Ronny Trommer
Christian Pape
Sebastian Rieger
Thomas Glotzbach
Keywords: Private Cloud; Energy Efficiency; Computer Networks; Power Management; OpenStack; AEQUO
Abstract:
Today's data centers need a huge amount of energy for their operation. Private cloud infrastructures using virtualization technologies are the prevailing paradigm in modern data centers and their energy consumption and the corresponding ongoing operational costs are not negligible. Solutions that raise the energy efficiency allow reductions in these operational costs and optimizations of the utilization of the data center infrastructure. Also, renewable energy sources can help to provide the needed energy, but usually these sources are fluctuating. Therefore, the energy is not always available when needed and also not always produced near the point of use. Further, the storage of energy is not available in industrial scale. The following article examines the possibility to shift the energy consumption of virtual machines and presents a lightweight prototype that can be integrated in private cloud environments using standard OpenStack components and application programming interfaces. It optimizes the energy efficiency by observing the current utilization parameters of compute resources and by taking appropriate actions based on this data. Furthermore, we evaluated mechanisms to control the energy efficiency of network resources. This optimization will be carried out by an automated instance, possessing a comprehensive view on the data center assets, which relocates virtual machines and optimizes the network structure. The article is completed with an evaluation to measure power consumption of data center assets during virtual machine live-migration operations and also illustrates further areas of research.
Pages: 13 to 28
Copyright: Copyright (c) to authors, 2015. Used with permission.
Publication date: June 30, 2015
Published in: journal
ISSN: 1942-2652