Optical Switching and Networking最新文献

We present a Demand-aware Reconfigurable Data Center Network architecture design (DROAD) with integrated fast-switching optics and space switches that allows dynamic reconfiguration and separation of intra- and inter-cluster connections. The performance analysis results show a 64% improvement in average Flow Completion Time and a significant reduction in TCP session time, as well as a reduced number of sessions needed to be opened compared to traditional electrically-switched leaf-spine networks.

With the prevalence of some high bandwidth-demanding applications, such as cloud computing, traditional wavelength-division-multiplexing passive optical networks have difficulties in satisfying such growing bandwidth demands due to its limited allocation-flexibility and utilization-efficiency. Therefore, elastic optical networks (EONs). In order to realize the flexibility in EONs, sophisticated routing and spectrum allocation (RSA) algorithms areone of the keyenabling technologies. However, most of the previous RSA algorithms were proposed with invariant routing and spectrum allocation strategies, which ignored considering the time-varying characteristics of EONs due to the variable network architecture and service provisioning. And such time-varying characteristics can deteriorate the spectrum fragmentation and the service blocking performances of EONs, which stimulates the application of various machine-learning technologies in EONs. In this paper, a long short-term memory based routing and spectrum assignment (LSTM-RSA) algorithm is proposed for EONs. By employing the long short-term memory technique to sense the complex status of EONs (e.g. spectral usage on the selected paths), the proposed LSTM-RSA algorithm gradually learns successful strategies through accumulating operation experience in the process of interaction and obtains higher returns through enhanced operation, which helps improve the spectrum fragmentation and the service blocking performances in EONs. Simulation results show that the spectrum fragmentation rate and the blocking rate of the proposed LSTM-RSA algorithm are reduced by about 6% and 8.9%, respectively, when compared to the traditional shortest-path-routing first-fitting RSA algorithm.

Telecom operators' infrastructure is sustained by optical communication networks that provide the means for exchanging large amounts of information, which is essential for many modern society needs. Optical networks are characterized by rapid breakthroughs in a variety of technologies. Relevantly, the last decade encompassed remarkable advances in optical networks’ subfields of signal processing, electronics, photonics, communications, protocols, and control-plane architectures. Hence, these advancements unlocked unprecedented transmission capacities, reconfigurability and programmability, entailing an evolution in the way which networks were designed, planned, and analyzed. In this paper, we review the historical status of optical planning and design tools by focusing on the major enabling technologies and relevant landmarks of the last decade(s). We begin by pinpointing the major breakthroughs in the optical data plane, estimation models capturing the transmission medium behavior and the control plane. We then distil the implications that these advancements entail in the landscape of optical network design and analysis tools, which commonly sit “on top” of the control plane or as a fully separated entity. Then, we speculate with our view for the future, in which automatic validation of optical network operations and dimensioning jointly with learning/artificial intelligence mechanisms will permit zero-touch optical networking: i.e. updating, provisioning, and upgrading network capacities, by means of automation with minimal human intervention. We conclude with a proposal of an architecture that encompasses data and control planes in a comprehensive manner for paving the way towards zero-touch optical networking.

The survivable logical topology mapping (SLTM) problem in IP-over-WDM networks is to map each link in the logical topology (IP layer) onto a lightpath in the physical topology (optical layer) such that a failure of a physical link does not cause the logical topology to become disconnected. This problem is known to be NP-complete. For this SLTM problem, two lines of investigations have been reported in the literature: the mathematical programming approach [1] and the structural approach introduced by Kurant and Thiran in [2] and pursued by Thulasiraman et al. [3,4,5]. In this paper we present an integrated treatment of the theoretical foundation of the survivable topology mapping problem presented in [3,4,5]. We believe that the algorithmic strategy developed in this paper will serve as an important phase in any strategy in the emerging area of resilient slicing of elastic optical networks. We conclude with a comparative evaluation, based on simulations, of the different algorithmic strategies developed in the paper, and also pointing to applications beyond IP-over-WDM optical networks, in particular, survivable design of inter-dependent multi-layer cyber physical systems such as smart power grids.

In today's wide area networks, especially in Optical Transport Networks (OTN) with Software Defined Networking (SDN) features enabled over Wavelength Division Multiplexing (WDM), Bandwidth on Demand (BoD) is an important service that can be satisfied by dynamic end-to-end service provisioning. Service provisioning time (SPT) and Blocking Probability (BP) are critical performance metrics for the users and carriers. This paper extends the concept of the Resource Delayed Release (RDR) strategy for WDM networks. The basic idea of this strategy is to introduce a delay in releasing the optical channel, when the channel is no longer carrying any services. This delay can help speed up the provisioning time for carrying the next service request, avoiding the time usually taken to establish a new optical channel. The main goals of the RDR method are to reduce SPT and BP while simultaneously satisfying the quality of service (QoS) constraints. In this paper, we investigate the effects of uniform and non-uniform traffic on the performance of RDR strategy. For non-uniform traffic simulation, we use a mesh topology with the 14 most populous cities in USA as of 2018 and model the non-uniform traffic based on population density. Further, we introduce a new metric called the Bandwidth Blocking Probability (BBP) to measure the quality of the service offered by the network. Simulation results show advantages of using the RDR method under a wide variety of traffic scenarios for both uniform and non-uniform traffic distributions compared to the traditional method. RDR reduces SPT by 45–90% for uniform traffic and 41–75% for non-uniform traffic. RDR reduces BP by 35–85% for uniform traffic and 30–75% for non-uniform traffic. Additionally, RDR lowers BBP by 31–73% for uniform traffic and 29–68% for non-uniform traffic.

Elastic optical networks (EONs) are a promising solution for future high-speed optical communication, and multicasting in EONs can efficiently support many emerging services. Different schemes, such as path, tree and subtree schemes, serve multicast services. In this paper, we consider a three-stage wavelength-space-wavelength (WSW) node architecture, which adopts wavelength switches in the first and last stages and space switches in the middle stage, and uses the path scheme to accommodate multicast requests, as proposed in an earlier work for elastic optical networks. We also enhance the WSW architecture to serve multicast requests in a more spectrum-efficient way, namely, using the subtree scheme, by making each switch support multicast capacity, and we term the resulting architecture M-WSW. To the best of our knowledge, this is the first study of the WSW architecture using the subtree scheme to support multicast capacity. We prove the sufficient and necessary conditions, in terms of the number of middle switches, of the M-WSW architecture for being strictly nonblocking (SNB) and wide-sense nonblocking (WSNB) under the two routing algorithms proposed in this paper. Our results show that the number of middle switches required for the architecture to be WSNB under each of the two proposed routing algorithms is much less than the number of middle switches required for SNB, especially when the SNB results meet the boundary condition.

In this special issue devoted to the memory of Prof. Fabio Neri we would like to look back at the time of the international research projects where some of us collaborated with him. On the basis of our personal experience of the time and the current viewpoint, we will discuss why Optical Packet Switching (OPS) is a technology that never came to market in spite of the great amount of research that was devoted to it. Then we will explore how Elastic Optical Network came to the stage more recently, somewhat overcoming the OPS technical proposal both in the interest of the researchers as well as of the industry.

In the last twenty years, optical networks have witnessed recurrent changes in their management and control architecture. In this paper, we present a historical timeline and a future perspective of the evolution of optical network management and control deployed for Wavelength Switched Optical Networks (WSON), Elastic Optical Networks (EON) and (multilayer) Data Center Networks.

Early implementations of WSON envisaged a static and centralized provisioning approach supported by the Management Plane only. Gradually, the requirement of accommodating more network dynamicity in WSON, and later in EON, pushed the adoption of a distributed control, mostly supported by vendor-dependent implementations of the Generalized MultiProtocol Label Switching (GMPLS) protocol suite. The drawbacks of the fully distributed GMPLS-based control, such as resource contention, suboptimal resource usage, and complex computations (e.g., to account for physical layer constraints) showed the necessity to bring back some of the routing/provisioning functions to a centralized Path Computation Element (PCE) capable of accounting for e.g. physical impairments and interworking with GMPLS.

The centralized control then gained its momentum and brought a radical change in network control, through the separation of data and control plane introduced by the paradigm of Software Defined Networking (SDN). Such an approach has been gradually extended to optical network control.

The paper, eventually, presents the most advanced control techniques, namely the intent-based networking, the observe/decide/act state-based approach providing for autonomic optical network and the (closed-loop) zero-touch service management approach. Advanced traffic conditioning techniques are also detailed, namely the in-band telemetry and the exploitation of Programming Protocol-Independent Packet Processors (P4) language capabilities as well as solutions tailored for data center networks: all of them are still in a research stage and to be integrated within future optical network architectures.

Next Generation Ethernet Passive Optical Network (NG-EPON) is considered to be future prospective access technology that could help to achieve 100Gbps data rates. Wavelength bonding is a phenomenon that can help Optical Network Units (ONU) to enhance their transmission capabilities. Using wavelength bonding, an ONU could transmit on multiple wavelength channels in parallel. The ONUs can achieve data transmission rates ranging from 25Gbps to 100Gbps. In upstream direction, ONUs share different available channels in time-sharing manner to effectively utilize the resources in NG-EPON. Dynamic Wavelength & Bandwidth Allocation (DWBA) algorithm is required for efficient allocation of wavelength and bandwidth resources in upstream direction. DWBA plays a vital role to help ONUs for transmission on multiple channels simultaneously. When an ONU transmits on multiple channels, a frame-rearrangement problem would occur at the Optical Line Terminal (OLT). OLT suffers from an extra overhead of frame-rearrangement; as the received frames at OLT are not in proper sequence. DWBA can play a vital role in avoiding frame rearrangement overhead. We proposed a DWBA algorithm to avoid/minimize frame rearrangement problem and efficient bandwidth allocation in NG-EPON. Our proposed DWBA avoids and minimizes frame-rearrangement problem and provides efficient resource allocation. We comparatively analyzed and evaluated our proposed DWBA with the existing DWBA algorithm. The simulation results show that our proposed DWBA minimizes frame-rearrangement problem as compared to existing DWBA algorithms and proves to be more efficient based on average (end-to-end) delay and completion time.

In this work the spectrum and power allocation (SPA) trade-off in elastic optical network (EON) is discussed in terms of the residual margin and residual spectrum in real-time application, both terms refer to the normalization of the sum-power and sum-spectrum in the resource allocation, respectively. Realistic scenarios have been investigated using optical performance monitoring techniques to measure the quality of transmission (QoT). The SPA-EON problem is formulated and three algorithms finding improved performance-complexity trade-offs are proposed to solve it: i) an analytical method based on combinatorial optimization, namely SPA-CO algorithm, ensuring the optimal solution but with a high computational cost; ii) a sub-optimum low-complexity method based on distance adaptive transmission (DAT), namely SPA-DAT, and iii) an SPA algorithm based on the distributed Verhulst algorithm, namely SPA-V, which achieves good solutions under acceptable computational time. A bunch of metrics, including probability of success, sum-power, and allocated spectrum are evaluated for the three SPA algorithms. The SPA-V was proved to be promising in EON operation, achieving the best performance-complexity trade-off.