Green Optical Communications—Part II: Energy Limitations in Networks

This is Part II of a two-part paper that explores the fundamental limitations on energy consumption in optical communications. Part I covers energy consumption in optical transport. Part II explores the lower bound on energy consumption in optical switches and networks, analyzes the energy performance of a range of switching devices, and presents quantitative models of the lower bounds on energy consumption in these devices. These models are incorporated into a simple model of a global switched network and the lower bound on total network energy consumption is estimated. We compare the results of this bottom-up calculation of the lower bound on network energy with a previous top-down analysis of overall network energy consumption based on real-world data for state-of-the art equipment and “business-as-usual” forward projections. The present analysis confirms a previous finding in that in a global scale network, the energy consumption of the switching infrastructure is larger than the energy consumption of the transport infrastructure. We find that the theoretical lower bounds on transport energy identified in Part I and the switching energy in this paper are more than three orders of magnitude lower than predicted by a “business-as-usual” analysis. In this paper, we explore how the gap between the theoretical lower bounds on energy consumption and current trends in network energy efficiency can be closed. We argue that future research needs to focus on improving the energy efficiency of switching and on devising methods to reduce the quantity of switching infrastructure in the network. Further key strategies for reducing network energy consumption include developing of low-energy transport technologies, reducing the energy overheads associated with peripheral functions that are not central to the transport and switching of data, and reducing the energy consumption of the access network.