I. Introduction

Today’s optical networks are facing new challenges and opportunities. The Optical Networks and Systems (ONS) Symposium, held during the IEEE Global Communications Conference (GLOBECOM) in Singapore between December 4-8, 2017, aimed to address these issues. GLOBECOM is one of the IEEE Communications Society’s two flagship conferences dedicated to driving innovation in nearly every aspect of communications. The ONS Symposium is the premier event for the optical networks community to showcase the latest developments in all research areas related to optical networks and enabling systems.

This feature issue is based on extended revised versions of the best papers presented at GLOBECOM 2017 during the ONS Symposium. The top 11 papers, out of the 33 presented at the event, were selected for this Feature Issue. Out of them, 8 manuscripts were submitted in response to the call for papers. After a strict peer-review process, 7 manuscripts were selected for this feature issue. A brief summary of the accepted papers is provided next.

In “Embedding Dependence-Aware Service Function Chains,” by M. Jalalitabar et al., the authors investigate the problem of Dependence-Aware Service Function Chain (D_SFC) design and mapping. They study how to efficiently map user’s service requests over the physical network while taking into consideration the computing resource demand, function dependence of the Virtual Network Functions (VNFs), and bandwidth demand of the service request. To solve this problem, they propose an algorithm called Dependence-aware SFC embedding with Group Mapping (D_SFC_GM). Their performance assessment shows that D_SFC_GM significantly outperforms traditional approaches based on topological sorting and sequential embedding.

In “Joint Banding-Node Placement and Resource Allocation for Multigranular Elastic Optical Networks,” by J. Wu et al., the authors address the problem of routing, fiber, waveband, and spectrum assignment (RFBSA) in multi-fiber-based elastic optical networks (EONs) with flexible waveband cross-connect nodes. They develop an integer linear programming formulation and a heuristic algorithm to solve this joint problem efficiently. Their results show that the proposed algorithm achieves good network performance, measured in terms of average maximum spectrum usage (MSU), as well as considerable hardware cost savings.

In “Joint Resource Allocation and Software-Based Reconfiguration for Energy-Efficient OFDMA-PONS,” by X. Gong et al., the authors investigate the problem of how to allocate SubCarriers (SCs) for energy-efficient upstream transmission in orthogonal frequency division multiplexing access passive optical networks (OFDMA-PONs). They leverage intuition to use software-based optical network unit (ONU) reconfigurations to reallocate SCs among segments in such a way that a minimum number of them is used to transmit user data. The authors proposed a heuristic that implements this concept. They also assess the performance of the heuristic via extensive simulations.

In “Evaluation Study of a Proposed Hadoop for Data Center Networks Incorporating Optical Circuit Switches,” by X. Wang et al., the authors focus on hybrid data center network (DCN) architectures and present a comprehensive comparative evaluation of a modified version of Hadoop, called Hadoop for hybrid networks (HHN). HHN creates network traffic patterns tailored for the characteristics of optical circuit switches (OCS) in hybrid DCNs. The benefits of using HHN are compared against using the original Hadoop in conventional electrical packet switches (EPS)-only networks. Numerical results demonstrate that HHN has similar system-level and user-level performance as the original Hadoop, thus allowing for full exploitation of the power and cost savings enabled by hybrid DCN architectures.

In “Traffic Classification and Sifting to Improve TDM-EPON Fronthaul Upstream Efficiency,” by Y. Wu et al., the authors aim to improve the upstream transmission efficiency when time division multiplexing ethernet passive optical networks (TDM EPONs) are employed to transmit mobile fronthaul (MF) flows. They propose an enhanced TDM EPON architecture based on traffic classification [i.e., classifying upstream traffic into useful and useless by using machine-learning (ML) classifiers] and useless-data sifting (i.e., sifting useless traffic to avoid the transmission of unnecessary EPON frames). The combination of these two techniques leads to significant improvements in terms of per-remote-radio-head (RRH) traffic load, number of RRHs supported by the same EPON, and signal-to-noise ratio (SNR), while keeping the end-to-end delay under 100 μs.

In “Benefits of Unidirectional Design Based on Decoupled Transmitters and Receivers in Tackling Traffic Asymmetry for Elastic Optical Networks,” by Y. Sheng et al., the authors tackle the traffic asymmetry problems in elastic optical networks (EONs) by decoupling a bidirectional transponder into a pair of separate unidirectional transmitters (Tx) and unidirectional receivers (Rx). The authors then consider different multi-flow transmitters which either use an array of laser diodes or a broadband laser source with filters to generate multiple sub-carriers. To evaluate the benefit of this unidirectional design, the authors solve the routing and spectrum allocation (RSA) optimization problem by means of an integer linear programming (ILP) model and an efficient heuristic algorithm based on spectrum window planes (SWPs). They show that their unidirectional design approach can significantly improve capacity utilization and minimize network cost.

In “Finding Survivable Routes in Multi-Domain Optical Networks With Geographically Correlated Failures,” by R. Gour et al., the authors address the problem of survivable path pair routing in multi-domain optical networks with geographically correlated failures. Their objective is to minimize the risk of simultaneous failures affecting both the primary and backup paths. To this end, they develop topology aggregation techniques and an inter-domain minimum overlapping area routing algorithm based on the aggregated information from each domain. Their proposed solution is shown to be effective in lowering the probability of having simultaneous failures.

We would like to thank all the authors who contributed to this feature issue. A special thank you also goes to all the reviewers for their effort and comments which greatly improved the quality of papers. We also thank the JOCN Editors-in-Chief for giving us the opportunity to prepare this feature issue and all the OSA publishing staff for their precious support.