Modeling Off-line Routing and Spectrum Allocation Problem in Elastic Optical Network

Abstract

The swift escalation in internet traffic due to diverse bandwidth starving applications and innovative concepts of modern technologies such as Elastic Optical Networks (EONs) and Software-defined networking (SDN) demands a dynamic and flexible optical network architecture both at the control and data plane. Characteristically, the flexibility in EONs is achieved by the emerging SDN-enabled sliceable bandwidth variable transponders (SBVTs) that support multiple optical carriers' simultaneous generation. These generated multiple optical carriers can operate different lightpaths using slice-ability or combined into a single high-rate super-channel. In this perspective, one of the major issues in EON is Routing and Spectrum Allocation (RSA). Typically, in EON, RSA is a spectrum management and Nondeterministic Polynomial-time hardness (NP-hard) problem in which network resources mainly bank on the applied ordering strategy. This article proposed a novel heuristic algorithm, Minimum Hops with Least Slot Spectrum (MHLS), to accommodate maximum traffic requests with better spectrum utilization. The proposed algorithm aims to minimize block requests, block traffic, and the total number of spectrum slots used in the network. The MHLS exploits Dijkstra-shortest-path and SDN-enabled SBVTs for RSA problem. The performance evaluation of MHLS is accomplished on the entire USA network.