Energy Consumption
The increase in the traffic in carried by optical networks will increase the energy consumption. The elastic optical network has the ability to significantly reduce energy consumption. In combination with sliceable bandwidth-variable transponders (SBVTs), the elastic optical network presents some new features as regards optical traffic grooming and optical layer bypass [58], which can help to reduce the energy consumption. However, we still are unable to groom traffic optically in a very early stage, and this omission must be tackled.
The elastic optical network offers a lower blocking probability compared to traditional optical networks and so can accept higher volumes of traffic. This clearly is a significant advantage in terms of energy efficiency, as the deployment of additional network elements would not only increase cost, but also increase the overall energy consumption. Several energy saving schemes are anticipated, setting some the network elements into sleep mode when the traffic is below a certain threshold. Another interesting topic for future researchers is to analyze the energy efficiency of new protection and restoration schemes for the elastic optical network.
Physical Layer Impairments
As optical connections may span over many long links, physical layer impairments (PLIs), such as—dispersion, interference, noise, and nonlinear effects accumulate and degrade the signal quality, which affects the quality-oftransmission (QoT). Accounting for PLIs is a challenging issue for network designers, especially if we consider exact models and the interdependencies. Many studies [87,273-275] on PLIs have been carried out for wavelength division multiplexing (WDM) based optical networks. PLIs have a distinct impact on both WDM-based optical networks and elastic optical networks. With the introduction of coherent detection and digital signal processing, impairments that are related to dispersion can be substantially reduced or fully compensated. However, high levels of flexibility make the minimization of these effects more complicated from an algorithmic perspective, which needs further research.
