Power-aware provisioning strategy with shared path protection in optical WDM networks |
| |
Authors: | Ning-Hai Bao Le-Min Li Hong-Fang Yu Zhi-Zhong Zhang Hong-Bin Luo |
| |
Affiliation: | 1. Key Lab of Broadband Optical Fiber Transmission and Communication Networks, University of Electronic Science and Technology of China, Chengdu 610054, PR China;2. Chongqing University of Posts and Telecommunications, Chongqing 400065, PR China;3. Beijing Jiaotong University, Beijing 100044, PR China |
| |
Abstract: | As the Internet continues to grow, the power consumption of telecommunication networks is rising at a considerable speed, which seriously increases the operational expenditure and greenhouse gas emission. Since optical Wavelength Division Multiplexing (WDM) networks are currently the most promising network infrastructures, power saving issue on these networks has received more attention in recent years. In traditional optical WDM networks, a large amount of power is drained by the redundant idle resources and reserved backup resources although these powered on resources do not carry traffic in most of the time. In order to reduce the network power consumption, turning off the corresponding network components or switching them to a low-power, standby state (or called sleep mode) is a promising greening approach. In this paper, we study the power-aware provisioning strategies and propose a sleep mode based Power-Aware Shared Path Protection (PASPP) heuristic algorithm to achieve the power efficiency of optical WDM networks. By jointly utilizing link-cost and fiber-cost in path routing, resource assignment, and resource release, PASPP makes working paths and backup paths converge on different fibers as much as possible, and switch idle and backup components to sleep mode to realize power saving. Simulation results show that our PASPP can obtain notable power saving and achieve satisfactory tradeoff between power efficiency and blocking probability with respect to Power-Unaware Shared Path Protection (PUSPP). |
| |
Keywords: | |
本文献已被 ScienceDirect 等数据库收录! |
|