Intelligent Resource Allocation Strategies for Next Generation Optical Networks

在互联网快速发展的推动下，光网络正在经历重大变革。下一代光网络基础设施将需要更加可定制和响应用户/应用需求。关键光传输技术的最新进展，如灵活的波分复用（WDM）网格，对流量异构的支持，以及软件定义光学（SDO）的出现，可以用来解决许多这些挑战。在未来，软件将根据带宽、传输质量（QoT）、能耗以及不同应用的可靠性和安全性需求等广泛参数，在控制网络组件和分配资源方面发挥越来越重要的作用。这将产生对开发复杂和自适应的资源分配策略和健壮的控制机制的巨大需求，这些策略和机制可以部署在下一代光网络基础设施的控制和管理平面中。

Optical networks are undergoing significant changes, fueled by the rapid growth of the internet. The next generation optical network infrastructure will need to be made more customizable and responsive to user/application requirements. Recent advances in key optical transport technologies such as flexible wavelength division multiplexing (WDM) grid, support for traffic heterogeneity, as well as the emergence of software defined optics (SDO) can be used to address many of these challenges. In the future, software will play an increasingly important role in controlling network components and allocating resources based on a wide range of parameters such as bandwidth, quality of transmission (QoT), energy consumption, as well as reliability and security needs of different applications. This will give rise to a tremendous need for developing sophisticated and adaptive resource allocation strategies and robust control mechanisms that can be deployed in the control and management plane of next generation optical network infrastructure. This research will focus on the development of innovative mechanisms for customizable resource allocation, and recovery from component failures as well as protecting data from malicious attacks. We will apply novel Operations Research (OR) techniques that exploit the inherent special structure of certain design problems to obtain optimal or near-optimal solutions. Recent work in this area has yielded promising results for network optimization; however, the full potential for the application of such methodologies is yet to be investigated properly. In addition to optimizing traditional network resources such as computing/processing power of nodes and capacity/bandwidth of the links, specialized algorithms to meet the following design objectives will be developed. 1. Intelligent scheduling techniques for energy efficient optical network design. 2. Routing and wavelength allocation strategies that minimize the risk of service disruption due to signal interference and malicious attacks. 3. Enhanced fault tolerance and survivability schemes for reliable communication in optical grids. 4. Optimized energy and bandwidth allocation for mixed line rate (MLR) networks.