Wide-area Optical Networks Research Articles

ILP-based resource optimization realized by quantum annealing for optical wide-area communication networks—A framework for solving combinatorial problems of a real-world application by quantum annealing

Resource allocation of wide-area internet networks is inherently a combinatorial optimization problem that if solved quickly, could provide near real-time adaptive control of internet-protocol traffic ensuring increased network efficacy and robustness, while minimizing energy requirements coming from power-hungry transceivers. In recent works we demonstrated how such a problem could be cast as a quadratic unconstrained binary optimization (QUBO) problem that can be embedded onto the D-Wave Advantage™ quantum annealer system, demonstrating proof of principle. Our initial studies left open the possibility for improvement of D-Wave solutions via judicious choices of system run parameters. Here we report on our investigations for optimizing these system parameters, and how we incorporate machine learning (ML) techniques to further improve on the quality of solutions. In particular, we use the Hamming distance to investigate correlations between various system-run parameters and solution vectors. We then apply a decision tree neural network (NN) to learn these correlations, with the goal of using the neural network to provide further guesses to solution vectors. We successfully implement this NN in a simple integer linear programming (ILP) example, demonstrating how the NN can fully map out the solution space that was not captured by D-Wave. We find, however, for the 3-node network problem the NN is not able to enhance the quality of space of solutions.

Accelerating model synchronization for distributed machine learning in an optical wide area network

Geo-distributed machine learning (Geo-DML) adopts a hierarchical training architecture that includes local model synchronization within the data center and global model synchronization (GMS) across data centers. However, the scarce and heterogeneous wide area network (WAN) bandwidth can become the bottleneck of training performance. An intelligent optical device (i.e., reconfigurable optical all-drop multiplexer) makes the modern WAN topology reconfigurable, which has been ignored by most approaches to speed up Geo-DML training. Therefore, in this paper, we study scheduling algorithms to accelerate model synchronization for Geo-DML training with consideration of the reconfigurable optical WAN topology. Specifically, we use an aggregation tree for each Geo-DML training job, which helps to reduce model synchronization communication overhead across the WAN, and propose two efficient algorithms to accelerate GMS for Geo-DML: MOptree, a model-based algorithm for single job scheduling, and MMOptree for multiple job scheduling, aiming to reconfigure the WAN topology and trees by reassigning wavelengths on each fiber. Based on the current WAN topology and job information, mathematical models are built to guide the topology reconstruction, wavelength, and bandwidth allocation for each edge of the trees. The simulation results show that MOptree completes the GMS stage up to 56.16% on average faster than the traditional tree without optical-layer reconfiguration, and MMOptree achieves up to 54.6% less weighted GMS time.

Multi-channel QoT prediction method in wide-area optical backbone network based on ensemble learning

Multi-channel QoT prediction method in wide-area optical backbone network based on ensemble learning

Job scheduling for distributed machine learning in optical WAN

Large companies operate tens of data centers (DCs) across the globe to serve their customers and store data. On the other hand, many machine learning applications need a global view of such global data to pursue high model accuracy. However, for this Geo-distributed machine learning (Geo-DML), it is infeasible to move all data together over wide-area networks (WANs) due to scarce WAN bandwidth, privacy concerns and data sovereignty laws. Therefore, most Geo-DML systems leverage geo-distributed approaches to train models, where global model synchronization is required between DCs over WAN. With the rapid increase of training data and the model sizes, it is challenging to efficiently utilize scarce and heterogeneous WAN bandwidth to synchronize models. With the advancement of optical technology, network topology becomes reconfigurable in optical WAN, which brings a new opportunity for Geo-DML training over WAN.We propose to optimize Geo-DML training with centralized joint control of the network and reconfigurable optical layers. We respectively prove the intra-job and inter-job scheduling problems are NP-hard and strongly NP-hard. For intra-job scheduling, RoWAN based on deterministic rounding algorithm, is presented to dynamically change the topology by reconfiguring the optical devices, and allocate path and rate for each flow. For inter-job scheduling, delayed SWRT is provided to schedule multiple jobs according to their priorities. The simulations in real topologies show that RoWAN reduces global model synchronization communication time of single iteration by up to 15.54%-48.2% on average in comparison with the traditional solutions. Compared to other three inter-job scheduling approaches, delayed SWRT can reduce the weighted job completion time (WJCT) by about 60%, 44.8% and 28.76%.

Deadline-Aware Multicast Transfers in Software-Defined Optical Wide-Area Networks

The increasing amount of data replication across datacenters introduces a need for efficient bulk data transfer protocols which provide certain guarantees, most notably timely transfer completion. We present DaRTree which leverages emerging optical reconfiguration technologies, to jointly optimize topology and multicast transfers in software-defined optical Wide-Area Networks (WANs), and thereby maximize throughput and acceptance ratio of transfer requests subject to transfer deadlines. DaRTree is based on a novel integer linear program relaxation and deterministic rounding scheme. To this end, DaRTree uses Steiner trees for forwarding and adaptive routing based on the current network load. DaRTree provides transfer completion guarantees without the need for rescheduling or preemption. Our evaluations show that DaRTree increases the network throughput and the number of accepted requests by up to 1.7×, especially for larger WANs. Moreover, DaRTree even outperforms state-of-the-art solutions when the traffic demands are only unicast transfers or when the WAN topology cannot be reconfigured. While DaRTree determines the rate and route to serve a request at the time of (online) admission control, we show that the acceptance ratio and throughput can be improved by up to 1.3× even further when DaRTree updates the rate and route of admitted transfers also at runtime.

BER-Adaptive RMLSA Algorithm for Wide-Area Flexible Optical Networks

Wide-area optical networks face significant transmission challenges due to the relentless growth of bandwidth demands experienced nowadays. Network operators must consider the relationship between modulation format and maximum reach for each connection request due to the accumulation of physical layer impairments in optical fiber links, to guarantee a minimum quality of service (QoS) and quality of transmission (QoT) to all connection requests. In this work, we present a BER-adaptive solution to solve the routing, modulation format, and spectrum assignment (RMLSA) problem for wide-area elastic optical networks. Our main goal is to maximize successful connection requests in wide-area networks while choosing modulation formats with the highest efficiency possible. Consequently, our technique uses an adaptive bit-error-rate (BER) threshold to achieve communication with the best QoT in the most efficient manner, using the strictest BER value and the modulation format with the smallest bandwidth possible. Additionally, the proposed algorithm relies on 3R regeneration devices to enable long-distances communications if transparent communication cannot be achieved. We assessed our method through simulations for various network conditions, such as the number of regenerators per node, traffic load per user, and BER threshold values. In a scenario without regenerators, the BER-Adaptive algorithm performs similarly to the most relaxed fixed BER threshold studied in blocking probability. However, it ensures a higher QoT to most of the connection requests. The proposed algorithm thrives with the use of regenerators, showing the best performance among the studied solutions, enabling long-distance communications with a high QoT and low blocking probability.

A Measurement Plane for Optical Networks to Manage Emergency Events

ABSTRACTIn this work, we show a wide geographical area optical network test bed, adopting the mPlane measurement plane for monitoring its performance and to manage software defined network approaches, with some specific tests and procedures dedicated to respond to disaster events and to support emergency networks. Such a test bed includes FTTX accesses, and it is currently implemented to support future 5G wireless services with slicing procedures based on Carrier Ethernet. The characteristics of this platform have been experimentally tested in the case of a damage-causing link failure and traffic congestion, showing a fast reactions to these disastrous events, allowing the user to recharge the initial QoS parameters.

Reliable Communication Networks for Smart Grid Transmission Systems

In this paper, we discuss the communications reliability requirements posed by the smart power grid with a focus on communications in support of wide area situational awareness. Implementation of wide area situational awareness relies on both transmission substation networks and wide area optical networks. We study the reliability of a sample communications network of the California Power Grid and find that its reliability falls short of proposed requirements. To overcome this issue, we consider the problem of designing the substation network and the wide area network to meet the reliability requirements while minimizing the network cost. For the wide area network design problem, we propose two alternate design approaches, namely: (1) following the power lines and (2) a mesh based design interconnecting the nodes. For the first approach we develop two greedy iterative heuristics and a heuristic integer linear programming (H-ILP) model using minimum cut-sets for network reliability optimization. The greedy iterative algorithms outperform the H-ILP approach in terms of cost, but require a larger amount of computing resources. Both proposed models are in fact complementary and thus provide a framework to optimize the reliability of smart grid communications networks restricted to following the power lines. In the second approach a greenfield mesh network method is proposed based on starting with a minimum spanning tree which is then augmented through a greedy heuristic into a mesh. Comparative numerical results show that the reliable mesh design has advantages in terms of the number of links and total link distance needed.

Design and Parametric Study of Depressed Core Optical Fiber

In this paper, the bending loss and the modal field distribution (MFD) of the depressed core optical fiber (DCOF) are investigated. The effect of optical and geometrical parameters on the bending loss, power confinement and MFD are examined in these fibers. Based on unique three layered structure of DCOF, the central air core, germanium silicate ring core, and silica cladding along with its mode transformation capability, we also demonstrated versatile applications in local area, wide area optical communication networks and tunable wavelength selective devices. Detailed design parameters and operation principles of the DCOF are discussed.

Regional Failure-Resilient Virtual Infrastructure Mapping in a Federated Computing and Networking System

Network-wide virtualization is a promising technique that will enable future Internet to support a variety of network services and architectures over a shared optical network substrate. In network-wide virtualization, it is important to efficiently map a virtual infrastructure (VI) over a wide-area optical network and guarantee the survivability of such a VI against failures. In this paper, we study the regional failure-resilient VI mapping (RRVIM) problem, where a failure-dependent protection approach is used to enable efficient sharing of protection resources among different regional failure scenarios. We first formulate the minimum-cost RRVIM problem using mixed integer linear programming (MILP) and then propose four heuristic algorithms to solve the problem. The efficiency of the MILP and the heuristics is compared via extensive simulation experiments under various performance metrics.

Hierarchical protection control system of smart substations

This paper analyzes the problems of the relay protections and secondary circuits of traditional substations, and introduces the hierarchical protection and control system of smart substations based on the technologies like IEC 61850 Standard, electronic instrument transformer, process level network, and wide area optical fiber communication network, etc. The hierarchical protection and control system includes three levels: local area protection, substation area protection and control, and wide area protection and control. These three levels are combined to solve the problems of the relay protections in traditional substations, thus improving protection performance and control capability of the system.

Energy-Efficient Resource-Provisioning Algorithms for Optical Clouds

Rising energy costs and climate change have led to an increased concern for energy efficiency (EE). As information and communication technology is responsible for about 4% of total energy consumption worldwide, it is essential to devise policies aimed at reducing it. In this paper, we propose a routing and scheduling algorithm for a cloud architecture that targets minimal total energy consumption by enabling switching off unused network and/or information technology (IT) resources, exploiting the cloud-specific anycast principle. A detailed energy model for the entire cloud infrastructure comprising a wide-area optical network and IT resources is provided. This model is used to make a single-step decision on which IT end points to use for a given request, including the routing of the network connection toward these end points. Our simulations quantitatively assess the EE algorithm’s potential energy savings but also assess the influence this may have on traditional quality-of-service parameters such as service blocking. Furthermore, we compare the one-step scheduling with traditional scheduling and routing schemes, which calculate the resource provisioning in a two-step approach (selecting first the destination IT end point and subsequently using unicast routing toward it). We show that depending on the offered infrastructure load, our proposed one-step calculation considerably lowers the total energy consumption (reduction up to 50%) compared to the traditional iterative scheduling and routing, especially in low- to medium-load scenarios, without any significant increase in the service blocking.

Building a Low-Energy Transparent Optical Wide Area Network With “Multipaths”

We propose an all-optical networking solution for a wide area network built on shared multipoint-to-multipoint lightpaths that, for short, we call “multipaths.” A multipath concentrates the traffic of a group of source nodes on a wavelength channel using an adapted medium access control protocol and multicasts this traffic to a group of destination nodes that extract their own data from the confluent stream. The proposed network can be built using existing components and appears less complex and more efficient in terms of energy consumption than alternatives like optical packet switching and optical burst switching. The paper presents the multipath architecture and compares its energy consumption to that of a classical router-based Internet service provider network. A flow-aware dynamic bandwidth allocation algorithm is proposed and shown to have excellent performance in terms of throughput and delay.

Performance analysis of fiber optical wide area network using SDM/WDM router

A spatial division multiplexing/wavelength division multiplexing (SDM/WDM) router has been designed for a fiber optical wide area network (WAN). The router includes a 32×32 optical switch and eight 1×4 demultiplexers (DeMUXs) and eight 4×1 multiplexers (MUXs). The system performances for various spans and transmission speeds are simulated and compared with the results of analytical calculations. Both the LiNbO3 and semiconductor optical amplifier (SOA) switches for the dilated Benes switch configuration are studied. The simulation results show that the SDM/WDM WAN using the SOA based router performs better than the LiNbO3.

A Cost Efficient Design of Virtual Infrastructures with Joint Node and Link Mapping

The virtualization of both servers and substrate networks is a promising technique that will enable the future Internet architecture to support a variety of cloud computing services and architectures. The problem of efficiently mapping a virtual infrastructure (VI) over a wide-area optical network is a key problem in provisioning Infrastructure (and Network) as a Service over multiple datacenters. In this paper, we study the problem of cost efficient VI mapping and propose a novel approach called the virtual infrastructure mapping algorithm (VIMA) that jointly considers node mapping and link mapping. Such a close coordination between the link and node mapping stages increases the efficiency of the VIMA algorithm. We compare the performance of our VIMA algorithm with other VI mapping algorithms such as NSVIM, vnmFlib and R-ViNE under various performance metrics using simulation. Our simulation results and analysis show that the VIMA algorithm outperforms the other algorithms in terms of VI mapping costs and path length of VI links.

RWA scheme minimizing FWM effect in a wide area optical network

In WDM based fiber optic communication systems, nonlinearities deteriorate network performance. A channel wavelength assignment scheme is required which not only minimize the FWM effect but also the network delay produced during assignment process. Wavelength assignment schemes along with impairment constraint based routing (ICBR) is proposed which minimizes the effect of four wave mixing and network delay in WAN. Results have been obtained for 8 and 16 equally spaced channels and observed that for equally spaced channels, in comparison with adjacent-next channel assignment scheme the proposed channel assignment scheme gives improvement in the results in terms of lower power of FWM products and lower number of in-band FWM products. Unlike existing channel assignment schemes the algorithm is simple and there is no need to calculate the FWM effect for each channel before assigning it hence reduction in network delay. Calculations have been obtained for point to point multi-fiber link of 1000 km, when power launched per channel is 1 mW for amplified wavelength convertible WAN with amplifier placement of 50 km apart.

CAPEX/OPEX effective optical wide area network design

The focus of this paper is on the design of the so-called Optical Wide Area Networks (owans), i.e., optical networks that cover broad areas. Our objective is to investigate efficient owan network design, where demand provisioning takes full advantage of the nodal switching equipment and of the network interface platforms under asymmetric traffic. It involves granting all traffic requests while minimizing the network capital and operational expenses, throughout an optimal dimensioning of the nodal equipment, i.e., minimizing the number and the location of the network nodal equipment. The originality of our work is in the forethought and the investigation of these issues. We establish a mathematical model which makes use of large scale optimization tools and propose a column generation algorithm coupled with a rounding off heuristic in order to solve it efficiently. In our experiments, with different network and traffic instances, we show that a careful dimensioning and location of the nodal equipment can save up to 35% of capital expenses, and even more sometimes.

Fast Restoration Based on Alternative Wavelength Paths in a Wide Area Optical IP Network

In this article we describe an experimental investigation of IP network restoration based on wavelength recovery. We propose a procedure for metro and wide area gigabit Ethernet networks that allows us to route the wavelength in case of link failure to another existing link by exploiting wavelength division multiplexing in the fiber. Such a procedure is obtained by means of an optical switch that is managed by a loss-of-light signal that is generated by a router in case of link failure. Such a method has been tested in an IP network consisting of three core routers with optical gigabit Ethernet interfaces connected by means of 50-km-long single-mode fibers between Rome and Pomezia. Compared with other conventional restoration techniques, such as OSPF and MPLS, our method -in very fast (20 ms) and is compatible with real-time TV services and low-cost chips.

Comparison between objective and subjective measurements of quality of service over an Optical Wide Area network

AbstractThe introduction of intelligence for management and control of the quality of service (QoS) are key issues for the evolution of the next generation IP optical network based on Ethernet technology. A test bed of a core‐access network was implemented to investigate such issues and in this paper, the authors report measurements concerning the QoS of multimedia services for different traffic conditions. The network was implemented with a differentiated service over Multi‐Protocol Label Switch (MPLS) architecture; it consists of a real optical network based on core routers with optical Gigabit Ethernet (GBE) interfaces connected by means of long single mode fibres (about 50\,km), contained in an installed cable between Rome and Pomezia. The network is based on different access devices, but in this paper we only refer to an access based on Fibre To The building (FTTB) architecture. The QoS was investigated both in terms of network (or objective) tests that include packet loss, jitter, one way delay and throughput measurements and perceptive (or subjective) tests that conversely are based on the evaluation of user perception. The services under test mainly consisted of video streams that circulated in the network according to the switching operations based in the IP differentiated service over MPLS technique that allowed us to guarantee the QoS for some class of service, also in the presence of network overload. Particular attention was given to the correlation between objective and subjective measurements. Furthermore, in order to have a complete analysis about QoS of real networks, measurements were performed also in the presence of restoration operations for link failures; in particular we compare the results in the case of conventional IP network restoration with a procedure proposed by us and based on a link switching activated by the loss of signal command coming from the routers. Copyright © 2007 John Wiley & Sons, Ltd.

UDT: UDP-based data transfer for high-speed wide area networks

In this paper, we summarize our work on the UDT high performance data transport protocol over the past four years. UDT was designed to effectively utilize the rapidly emerging high-speed wide area optical networks. It is built on top of UDP with reliability control and congestion control, which makes it quite easy to install. The congestion control algorithm is the major internal functionality to enable UDT to effectively utilize high bandwidth. Meanwhile, we also implemented a set of APIs to support easy application implementation, including both reliable data streaming and partial reliable messaging. The original UDT library has also been extended to Composable UDT, which can support various congestion control algorithms. We will describe in detail the design and implementation of UDT, the UDT congestion control algorithm, Composable UDT, and the performance evaluation.