Proceedings of the 2003 conference on Applications, technologies, architectures, and protocols for computer communications
Green networks: energy efficient design for optical networks
WOCN'09 Proceedings of the Sixth international conference on Wireless and Optical Communications Networks
Energy saving and cost reduction in multi-granularity green optical networks
Computer Networks: The International Journal of Computer and Telecommunications Networking
Multicloud Deployment of Computing Clusters for Loosely Coupled MTC Applications
IEEE Transactions on Parallel and Distributed Systems
Energy Efficiency in Telecom Optical Networks
IEEE Communications Surveys & Tutorials
An energy-aware dynamic RWA framework for next-generation wavelength-routed networks
Computer Networks: The International Journal of Computer and Telecommunications Networking
Power consumption and energy efficiency in the internet
IEEE Network: The Magazine of Global Internetworking
Journal of High Speed Networks
Hi-index | 0.00 |
Over the past decade, the ever-increasing energy demands of IT infrastructures have posed significant challenges for the research community in terms of reducing their total power consumption and minimizing their environmental impact. Optical communication networks are envisioned to be promising candidates to help preventing this problem affecting the Internet backbone, as well as for distributed applications such as computational grids. In this paper, we propose an adaptive and distributed scheme for the establishment of energy-efficient lightpaths in computational grids. The grid is deployed over an optical circuit-switched backbone network, defining an optical grid network. Each node of the backbone network maintains two different dynamic thresholds values and estimates the changes in network performance by evaluating the moving average of the total wavelength channel occupancy on all its input/output links. The nodes have the ability of reducing the energy consumption by entering into an Energy Saving Mode ESM on the basis of a comparison between their channel occupancy and the thresholds. Furthermore, we extend our framework by allowing the thresholds to be dynamically adapted depending on the network performance in terms of blocking probability. We show that the proposed method achieves considerable energy savings when compared to a normal energy-unaware operational mode and still allows to maintain an acceptable level of network performance in terms of blocking probability and end-to-end delay. Numerical results are obtained with a Java event-driven simulator of two different optical network topologies.