Multi-stratum resources resilience in software defined data center interconnection based on IP over elastic optical networks

IP over elastic optical network is a very promising networking architecture to interconnect data centers. It can enable efficient resource utilization and support heterogeneous bandwidth demands in cost-effective, highly available, and energy-effective manner. In case of aggregation elastic optical network node failure, to ensure a high-level quality of service for user request after the failure becomes a research focus. In this paper, we present a novel multi-stratum resources resilience (MSRR) architecture for the data center services in software defined data center interconnection based on IP over elastic optical networks. The MSRR can enable joint optimization of IP network, elastic optical network, and application stratum resources, and enhance the service resilience and the data center responsiveness to the dynamic end-to-end service demands. Additionally, a service-aware resource collaborative resilience strategy for MSRR is introduced based on the proposed architecture, which can provide the restoration using the multiple stratums resources in case of failure. The overall feasibility and efficiency of the proposed architecture are experimentally verified on our testbed. Moreover, the network performances are quantitatively evaluated through the simulation under heavy traffic load scenario in terms of path blocking probability, resource occupation rate, and path resilience latency.     

Multicast service protection algorithm based on elastic optical network

With the rapid growth of the network traffic,the elastic optical network (EON) has been proposed as a promising solution due to its high spectrum efficiency and flexible bandwidth provision.Meanwhile,multicast routing and spectrum allocation,and the survivability of the network become more challenging than that in the conventional optical network.The routing for multicast traffic and its protection algorithm in EON was investigated.An integer linear programming (ILP) formulation with the objective to minimize total spectrum consumption was presented.In addition,a heuristic algorithm called multicast sub-tree protection algorithm (MSPA) to achieve sufficient protection and satisfy resources savings was designed.The simulation results demonstrate that comparing with the traditional multicast routing and protection algorithm,MSPA performs well in improving the blocking probability and the spectrum utilization of the network.     

Virtual infrastructure mapping in software-defined elastic optical networks

Software-defined networking (SDN) enables efficient and scalable network virtualization, which allows infrastructure resources such as computing and networking resources to be abstracted and outsourced as a service. The SDN technologies can be extended to the optical transport networks to achieve an intelligent and flexible resource management, thus achieving high-capacity, low-cost, and long-reach optical transport networks. In this paper, we introduce an architecture for software-defined elastic optical networks and study the virtual infrastructure (VI) mapping problem with the objective of minimizing the blocking probability. We propose a set of efficient heuristic algorithms, among which the Network followed by Compute Load Balancing (NCLB) algorithm is a novel attempt to solve the VI mapping problem by provisioning the networking resource first followed by allocating the computing resource. Furthermore, we propose a modified version of NCLB, called Network Depth-based NCLB (ND-NCLB), which confines the VI mapping assignment in a small-range sub-network to further optimize the physical network resource consumption. Through comprehensive simulations in various scenarios, we demonstrate that the proposed ND-NCLB algorithm achieves the best performance in terms of blocking probability compared to the other algorithms in this work.     

Time-sensitive software-defined networking (Ts-SDN) control architecture for flexi-grid optical networks with data center application

Data center interconnected by flexi-grid optical networks is a promising scenario to meet the high burstiness and high bandwidth requirement of data center application, because flexi-grid optical networks can allocate spectral resources for applications in a dynamic, tunable and efficient control manner. Meanwhile, as centralized control architecture, the software-defined networking (SDN) enabled by OpenFlow protocol can provide maximum flexibility for the networks and make a unified control over various resources for the joint optimization of data center and network resources. Time factor is firstly introduced into SDN-based control architecture for flexi-grid optical networks supporting data center application. A traffic model considering time factor is proposed, and a requirement parameter, i.e., bandwidth-period product is adopted for the service requirement measurement. Then, time-sensitive software-defined networking (Ts-SDN)-based control architecture is designed with OpenFlow protocol extension. A novel deadline-driven PCE algorithm is proposed for the deadline-driven service under Ts-SDN-based control architecture, which can complete data center selection, path computation and bandwidth resource allocation. Finally, simulation results shows that our proposed Ts-SDN control architecture and deadline-driven PCE algorithm can improve the application and network performance to a large extent in blocking probability.     

Algorithms for multicast traffic grooming in WDM mesh networks

Several of the new applications in high-performance networks are of the multicast traffic type. Since such networks employ an optical network infrastructure, and since most of these applications require subwavelength bandwidth, several streams are usually groomed on the same wavelength. This article presents an account of recent advances in the design of optical networks for multicast traffic grooming in WDM mesh networks. The article addresses network design and session provisioning under both static and dynamic multicast traffic. Under static traffic conditions, the objective is to accommodate a given set of multicast traffic demands, while minimizing the implementation cost. Optimal and heuristic solution techniques for mesh network topologies are presented. Under dynamic traffic conditions, techniques for dynamic routing and session provisioning of multicast sessions whose objective is to minimize session blocking probabilities are explained. The article also presents a number of open research issues     

Adaptive multilevel modulation for grooming in elastic cloud optical networks

In order to mitigate the mismatch of granularities between fixed grid and client traffic, the elastic optical network (EON) was proposed by using orthogonal frequency division multiplexing. In EONs, the bandwidth variable transponder adjusts the number of bits per symbol, so that an optical signal generates by using just enough sub-carriers each with appropriate modulation level. Owing to the advantage of line-rate adaption above, the application of cloud computing has witnessed rapid growth in EONs. However, bandwidth variable transponders consume more power compared with ordinary ones, which will lead to a power-thirsty EON if no effective measure is taken. As a result, the green grooming was proposed for EONs. Unfortunately, the adaptive multilevel modulation was neglected in the current works focusing on green grooming. Thus, in this paper, we design a novel modulation adaptive grooming with guaranteeing transmission performances in green EONs. The distance-adaptive spectrum resource allocation is applied to the green grooming algorithm previously designed by us for EONs. The simulation results show that the adaptive multilevel modulation plays an important role on saving spectrum and power consumption for the green grooming in EONs, because the spectral bandwidth can be saved by increasing the number of bits per symbol to transmit the same data rate.     

基于高阶循环累积量的OFDM-弹性光网络信号调制格式识别

将高阶循环累积量引入基于正交频分复用(OFDM)的弹性光网络(EON)的调制格式识别中。通 过计算OFDM信号不同调制格式(BPSK、QPSK、16QAM和64QAM)的高阶循环累积量,给 出了OFDM-EON信号不同调制格式的识别阈值区间,并采用仿真方法验证了该区间的 识别准确率。结果表明,利用所提出的识别阈值区间,对波特率为28G的OFDM-EON信号(各 子载波随机设置调制格式)进行调制格式识别,在光信噪比(OSNR)为34dB时,子载波调制格式的识别准确率可达到100%。     

Proactive defragmentation in elastic optical networks under dynamic load conditions

The main weakness of elastic optical networks (EON), under dynamic traffic conditions, stems from spectrum fragmentation. A lot of research efforts have been dedicated during recent years to spectrum defragmentation. In this work, a thorough study about proactive defragmentation is carried out. Effects of the different defragmentation parameters on the EON performance are analyzed, and appropriate values of the defragmentation period, which guarantee suitable network performance while keeping the network control complexity at reasonable values, are obtained by means of extensive simulations. Benefit obtained by applying different defragmentation strategies, in terms of increase in the supported load at a given bandwidth blocking probability, is also reported. Different traffic conditions and network topologies are simulated to assess the validity of the obtained results.     

Naive-LSTM based services awareness of edge computing elastic optical networks

Great challenges and demands are presented by increasing edge computing services for current elastic optical networks(EONs) to deal with serious diversity and complexity of these services. To improve the match degree between edge computing and optical network, the services awareness function is necessary for EON. This article proposes a Naive long short-term memory(Naive-LSTM) based services awareness strategy of the EON, where the Naive-LSTM model makes full use of the most simplified structure...     

Dynamic fragments awareness based virtual network mapping strategy of elastic optical networks

As virtual networks services emerge increasingly with higher diversification, the issue of spectrum fragments presents great challenge to the elastic optical networks (EON), especially under heaven services burdens. Aimed to solve this problem, this article proposes a dynamic fragments awareness based virtual network mapping (DFA-VNM) strategy of elastic optical network. In this proposed approach, the dynamic fragments awareness model of it is established, which takes available bandwidth demand and spectrum fragment degree into consideration. Moreover, the dynamic fragments awareness based virtual network mapping strategy makes full advantage of real-time fragments awareness result to conduct virtual network service mapping operation with less fragments and lower blocking rate. Testing results show that the proposed approach is able to improved services supporting ability of EON.     

Reliable Multipath Provisioning for High-Capacity Backbone Mesh Networks

We investigate reliable multipath provisioning of traffic in high-capacity backbone mesh networks, e.g., next-generation SONET/SDH networks supporting virtual concatenation (VCAT). VCAT enables a connection to be inversely multiplexed on to multiple paths, a feature that may lead to significantly improved performance over conventional single-path provisioning. Other mesh networks such as those employing optical wavelength-division multiplexing (WDM) and multiprotocol label switching (MPLS) can also benefit from this multipath provisioning approach. We propose effective multipath bandwidth as the metric to provision a connection while satisfying its reliability requirements (measured in terms of availability). We demonstrate that effective multipath bandwidth provides more flexibility and lower blocking probability without the cost and the complexity associated with traditional protection schemes developed for optical WDM and MPLS networks. We also investigate the practical problem of provisioning effective multipath bandwidth with cost constraints. We analyze the tractability of the problem and present a heuristic which results in significantly reduced number of blocked connections due to cost constraints.     

弹性光网络双重故障下的自适应保护级别算法

为了降低弹性光网络中双链路故障保护下的业务及带宽阻塞率、均衡带宽资源分配,提出了一种双重故障下的自适应保护级别算法。该算法综合考虑传输距离、调制格式等因素,根据链路频谱资源使用状态动态地更新链路惩罚系数,选取最佳路径进行传输,同时根据请求类型及网络空闲资源状态自适应地选择保护方法,为业务提供最大限度的保护。仿真结果表明,在双链路故障下,算法在阻塞率方面取得了较好的性能,同时均衡了网络中各链路上的资源使用。     

Reinforced virtual optical network embedding algorithm in EONs for edge computing

As the core technology of optical networks virtualization, virtual optical network embedding(VONE) enables multiple virtual network requests to share substrate elastic optical network(EON) resources simultaneously and hence has been applicated in edge computing scenarios. In this paper, we propose a reinforced virtual optical network embedding(R-VONE) algorithm based on deep reinforcement learning(DRL) to optimize network embedding policies automatically. The network resource attributes are extr...     

光纤网络中云计算资源联合最优分配方法

在云计算技术与光纤网络不断发展的今天,网络数据不断增加,网络带宽面临严峻挑战。当前应用云计算资源联合最优分配方法对网络链路资源分配的结果较为落后,导致资源分配后网络数据传输扰动量以及带宽阻塞率较高。针对此问题,本文设计光纤网络中云计算资源联合最优分配方法。以网络任务模型为蓝本,构建光纤网络传输模型。将光纤网络传输模型以有向图的形式,完成网络节点传输映射分析。结合光树生成算法,设计云计算资源最优分配算法,完成光纤网络中云计算资源联合最优分配方法设计。仿真结果表明：采用所提方法后网络数据传输扰动量以及带宽阻塞率较低,提升了云计算资源的分配方法。     

WDM网状网络中一种动态多播自适应业务疏导算法

光多播业务需要消耗大量的WDM光网络带宽资源,业务疏导是光多播网络降低业务请求阻塞率和提高带宽资源利用率的有效方法。提出了一种新型光多播疏导节点结构,研究了疏导端口优先的多播业务疏导算法(TGPFA)和新建光树优先的多播业务疏导算法(TCLFA),进而提出了一种能够适应网络资源变化的动态多播业务疏导算法(ADMGA)。结果表明,在网络资源有限的情况下,ADMGA算法能取得较低的请求阻塞率和带宽阻塞率,获得较好的网络性能。     

Software defined elastic optical networking in temporal,spectral, and spatial domains

This paper discusses architecture, protocol, technologies, systems, and networking testbed for software-defined elastic optical networking in temporal, spectral, and spatial domains. By exploiting the progress in elastic optical networking (EON) in temporal and spectral domains utilizing dynamic optical arbitrary waveform generation and measurement technologies, and by extending the EON concept into the spatial domain through the new orbital angular momentum-based spatial division multiplexing, we realize EON exploiting elasticity in temporal, spectral, and spatial domains (3D-EON). Routing, spectral, spatial mode, and modulation format assignment with fragmentation awareness as well as hitless defragmentation for high capacity, high quality of service, and resource-efficient networking will be pursued. OpenFlow-based 3D-EON testbed at UC Davis includes optical supervisory channel with optical performance monitoring for software-defined networking with an adaptive observe-analyze-act cycle.     

Improved Dynamic Lightpath Provisioning for Large Wavelength-Division Multiplexed Backbones

Technology already exists that would allow future optical networks to support automatic lightpath configuration in response to dynamic traffic demands. Given appropriate commercial drivers, it is possible to foresee carrier network operators migrating away from semipermanent provisioning to enable on-demand short-duration communications. However, with traditional lightpath reservation protocols, a portion of the lightpath is idly held during the signaling propagation phase, which can significantly reduce the lightpath bandwidth efficiency in large wavelength-division multiplexed backbones. This paper proposes a prebooking mechanism to improve the lightpath efficiency over traditional reactive two-way reservation protocols, consequently liberating network resources to support higher traffic loads. The prebooking mechanism predicts the time when the traffic will appear at the optical cross connects, and intelligently schedules the lightpath components such that resources are only consumed as necessary. We describe the proposed signaling procedure for both centralized and distributed control planes and analyze its performance. This paper also investigates the aggregated flow length characteristics with the self-similar incident traffic and examines the effects of traffic prediction on the blocking probability as well as the ability to support latency sensitive traffic in a wide-area environment.     

Traffic engineering in multigranularity heterogeneous optical WDM mesh networks through dynamic traffic grooming

In this article we investigate the problem of efficiently provisioning connections of different bandwidth granularities in a heterogeneous WDM mesh network through dynamic traffic grooming schemes under traffic engineering principles. Due to the huge amount of traffic a WDM backbone network can support and the large geographic area it can cover, constructing and upgrading such an optical WDM network can be costly. Hence, it is extremely important for network operators to apply traffic engineering strategies to cost-effectively support different bandwidth granularity services using only the appropriate amount of network resources. This requires an optical WDM network to have multigranularity switching capability, and such a network tends to be a multivendor heterogeneous network. However, WDM network heterogeneity increases the difficulty and challenge of efficient traffic provisioning. In this article we present different TE issues that need to be carefully considered in such an optical WDM network, and propose possible solutions and extensions for the generalized multiprotocol label switching optical network control plane. We extend an existing generic graph model to perform efficient traffic grooming and achieve different TE objectives through simple shortest path computation algorithms. We show that our approach is very practical and very suitable for traffic engineering in a heterogeneous multigranularity optical WDM mesh network.     

基于光森林的弹性光网络能效组播路由频谱分配策略

在弹性光网络中,光树传输组播可以节省链路代价,但较长的光树需要选择更低的调制等级,消耗更多的频谱资源和发射机功耗.提出一种基于遗传算法的光森林组播和光树重配置(GAMF-TR)优化组播的能效路由、调制格式和频谱分配(RMSA)策略.GAMF-TR策略设计染色体编码表示光森林的组播目的节点划分和光路径组合,通过染色体的基因位概率交叉和变异得到更多的光森林RMSA策略,设计了一个频谱分配效率和发射机功耗折中的适应度函数选择能效最高的光森林RMSA策略,并设计在网络资源充足时将组播从光森林重配置到光树传输,进一步减少发射机功耗的消耗.仿真结果表明:提出的GAMF-TR策略获得了组播的最低带宽阻塞率和最高能效性能.     

Virtual network mapping in elastic optical networks with sliceable transponders

Emerging inter-datacenter applications involving data transferred, processed, and analyzed at multiple datacenters, such as virtual machine migrations, real-time data backup, remote desktop, and virtual datacenters, can be modeled as virtual network requests that share computing and spectrum resources of a common substrate physical inter-datacenter network. Recent advances make elastic optical networks an ideal candidate for meeting the dynamic and heterogeneous connection demands between datacenters. In this paper, we address the static (offline) version of the virtual network mapping problem in elastic optical networks equipped with sliceable bandwidth variable transponders. The objective is to minimize the total transponder cost in the network. An Integer Linear Programming formulation is presented, lower bounds are derived, and six heuristics are proposed and compared. Simulation results are presented to show the effectiveness of the proposed approaches.