On Efficient Oblivious Wavelength Assignments for Programmable Wide-Area Topologies

Abstract

Given the explosively growing traffic related to data-centric applications and AI, especially to and from the cloud, it is crucial to make the best use of the given resources of wide-area backbone networks (WANs). An intriguing approach to improve both efficiency and performance of WANs is to render networks more adaptive and "demand-aware", on the physical layer: innovative programmable wide-area topologies support dynamic wavelength assignments. This is enabled by the application of colorless and directionless Reconfigurable Optical Add/Drop Multiplexers (CD ROADM), and by leveraging the capabilities of software-defined controllers. This paper investigates the benefit of such fully dynamic wavelength assignments in programmable WAN topologies, compared to an oblivious wavelength assignment. To this end, we also propose a new demand-oblivious strategy to optimize the capacity of a WAN. Considering both real and synthetic scenarios, we find that our proposed demand-oblivious strategy can perform close to dynamic approaches with respect to throughput, without entailing reconfiguration costs.