Historical data learning based dynamic LSP routing for overlay IP/MPLS over WDM networks

Overlay IP/MPLS over WDM network is a promising network architecture starting to gain wide deployments recently. A desirable feature of such a network is to achieve efficient routing with limited information exchanges between the IP/MPLS and the WDM layers. This paper studies dynamic label switched path (LSP) routing in the overlay IP/MPLS over WDM networks. To enhance network performance while maintaining its simplicity, we propose to learn from the historical data of lightpath setup costs maintained by the IP-layer integrated service provider (ISP) when making routing decisions. Using a novel historical data learning scheme for logical link cost estimation, we develop a new dynamic LSP routing method named Existing Link First (ELF) algorithm. Simulation results show that the proposed algorithm significantly outperforms the existing ones under different traffic loads, with either limited or unlimited numbers of optical ports. Effects of the number of candidate routes, add/drop ratio and the amount of historical data are also evaluated.     

Study on a joint multiple layer restoration scheme for IP over WDM networks

In this article we investigate the problem of a restoration scheme for IP over WDM networks. Network reliability is gaining importance with the huge volume of traffic carried by such networks. Providing survivability at the optical layer is inherently attractive, but raises many questions and challenges, given the characteristic of optical aggregated lightpath and relatively coarse traffic granularity. The emergence of MPLS and its extension, MP/spl lambda/S, opens up new possibilities for developing simple integrated protection/restoration schemes that can be coordinated at both the IP and optical layers . This article first presents an overview of existing MPLS/MP/spl lambda/S recovery mechanisms. Then we propose a joint two-layer recovery scheme for IP-centric WDM-based optical networks where the optical layer takes the recovery actions first, and subsequently the upper IP layer initiates its own recovery mechanism, if the optical layer does not restore all affected services. A simulation-based analysis shows the benefits of the proposed two-layer recovery scheme over single-layer recovery schemes. We demonstrate the advantages of finer granularity in IP layer recovery and the effectiveness in speed on the optical layer. The impact of several network parameters on recovery performance is also studied in the paper.     

Physical layer performance benchmarking of two metropolitan area network configurations

This paper investigates the problem of dynamic survivable lightpath provisioning against single-node/link failures in optical mesh networks employing wavelength-division multiplexing (WDM).We unify various forms of segment protection into generalized segment protection (GSP). In GSP, the working path of a lightpath is divided into multiple overlapping working segments, each of which is protected by a node-/link-disjoint backup segment. We design an efficient heuristic which, upon the arrival of a lightpath request, dynamically divides a judiciously selected working path into multiple overlapping working segments and computes a backup segment for each working segment while accommodating backup sharing. Compared to the widely considered shared-path protection scheme, GSP achieves much lower blocking probability and shorter protection-switching time for a small sacrifice in control and management overhead.On the basis of generalized segment protection, we present a new approach to provisioning lightpath requests according to their differentiated quality-of-protection (QoP) requirements. We focus on one of the most important QoP parameters—namely, protection-switching time—since lightpath requests may have differentiated protection-switching-time requirements. For example, lightpaths carrying voice traffic may require 50 ms protection-switching time while lightpaths carrying data traffic may have a wide range of protection-switching-time requirements. Numerical results show that our approach achieves significant performance gain which leads to a remarkable reduction in blocking probability.While our focus is on the optical WDM network, the basic ideas of our approaches can be applied to multi-protocol label switching (MPLS) networks with appropriate adjustments, e.g., differentiated bandwidth granularities.     

A Bandwidth Guaranteed Integrated Routing Algorithm in IP over WDM Optical Networks

In this paper, we have developed an integrated online algorithm for dynamic routing of bandwidth guaranteed label switched paths (LSPs) in IP-over-WDM optical networks. Traditionally, routing at an upper layer (e.g., IP layer) is independent of wavelength routing at the optical layer. Wavelength routing at the optical layer sets up a quasi-static logical topology which is then used at the IP layer for IP routing. The coarse-grain wavelength channels and the pre-determined virtual topologies with respect to some a priori assumed traffic distribution are barriers to efficient resource use and inflexible to changing traffic. We take into account the combined knowledge of resource and topology information at both IP and optical layers. With this added knowledge, an integrated routing approach may extract better network efficiencies, be more robust to changing traffic patterns at the IP layer than schemes that either use dynamic routing information at the IP layer or use a static wavelength topology only. LSP set-up requests are represented in terms of a pair of ingress and egress routers as well as its bandwidth requirement, and arrive one-by-one. There is no a priori knowledge regarding the arrivals and characteristics of future LSP set-up requests. Our proposed algorithm considers not only the importance of critical links, but also their relative importance to routing potential future LSP set-up requests by characterizing their normalized bandwidth contribution to routing future LSP requests with bandwidth requirements. Moreover, link residual bandwidth information that captures the link's capability of routing future LSPs is also incorporated into route calculation. Extensive simulation was conducted to study the performance of our proposed algorithm and to compare it with some existing ones, such as the integrated minimum hop routing algorithm and the maximum open capacity routing algorithm. Simulation results show that our proposed algorithm performs better than both routing algorithms in terms of the number of LSP set-up requests rejected and the total available bandwidth between router pairs.     

An efficient mechanism for dynamic multicast traffic grooming in overlay IP/MPLS over WDM networks

This paper proposes an efficient overlay multicast provisioning (OMP) mechanism for dynamic multicast traffic grooming in overlay IP/MPLS over WDM networks. To facilitate request provisioning, OMP jointly utilizes a data learning (DL) scheme on the IP/MPLS layer for logical link cost estimation, and a lightpath fragmentation (LPF) based method on the WDM layer for improving resource sharing in grooming process. Extensive simulations are carried out to evaluate the performance of OMP mechanism under different traffic loads, with either limited or unlimited port resources. Simulation results demonstrate that OMP significantly outperforms the existing methods. To evaluate the respective influences of the DL scheme and the LPF method on OMP performance, provisioning mechanisms only utilizing either the IP/MPLS layer DL scheme or the WDM layer LPF method are also devised. Comparison results show that both DL and LPF methods help improve OMP blocking performance, and contribution from the DL scheme is more significant when the fixed routing and first-fit wavelength assignment (RWA) strategy is adopted on the WDM layer. Effects of a few other factors, including definition of connection cost to be reported by the WDM layer to the IP/MPLS layer and WDM-layer routing method, on OMP performance are also evaluated.     

Protection Schemes for IP-over-WDM Networks: Throughput and Recovery Time Comparison

This article presents a novel protection approach using Generalized Multi-Protocol Label Switching (GMPLS). This strategy provides protection at the Wavelength Division Multiplexing (WDM) layer, meaning that all Internet Protocol Label-Switched Path (IP LSPs) nested inside a lightpath are protected in an aggregated way. It uses resources efficiently since spare capacity of working primary lightpaths can be used for backup purposes whenever necessary. The IP and WDM layers are treated together as a single integrated network from a control plane point of view, so that network state information from both layers can be used. Besides discussing the strategy proposed and the key features of GMPLS that will allow its implementation, we mathematically formulate the maximum throughput problem. Thereafter, we propose and compare heuristic algorithms for IP-over-WDM networks using three protection approaches: WDM lightpath protection, IP LSP protection, and the proposed protection scheme. Their throughputs and recovery times are analyzed and compared. Our results show that, for a representative mesh network, the proposed aggregated protection scheme presents better protection efficiency and good scalability properties when compared with the other two schemes.     

光网络中基于SRLG约束的动态RWA算法

在IPoverWDM光网络中,需要为每条工作光路提供光通路保护,即提供一条保护光路。光网络中引入了共享风险链路组(SRLG)的概念,要求工作光路与保护光路不共享风险链路组。不考虑波长分配提出一种新的启发式算法,该算法同时计算两条SRLG分离且总费用低;在此算法基础上利用分层图的思想,提出WDM网络中的一种在SRLG约束条件下的动态选路波长分配算法。     

智能光网中的联合路由

在IP/MPLS智能光网络中的LSP路由分为独立路由和联合路由两类。由于综合考虑了光层和IP层的可用资源信息和拓扑信息,联合路由能够提供比独立路由方案更高的资源利用率。文中提出了联合路由算法的设计目标,综述了联合路由算法的研究现状,并给出了下一步的研究方向。     

Efficient Distributed Bandwidth Management for MPLS Fast Reroute

As service providers move more applications to their IP/MPLS (multiple protocol label switching ) backbone networks, rapid restoration upon failure becomes more and more crucial. Recently MPLS fast reroute has attracted lots of attention as it was designed to meet the needs of real-time applications, such as voice over IP. MPLS fast reroute achieves rapid restoration by computing and signaling backup label switched path (LSP) tunnels in advance and re-directing traffic as close to failure point as possible. To provide a guarantee of bandwidth protection, extra bandwidth has to be reserved on backup paths. Using path merging technique as described in IETF RFC 4090 only, the network is able to share some bandwidth on common links among backup paths of the same service LSP, i.e., so-called intra-sharing. But no solution is provided on how to share bandwidth among backup paths of different service LSPs, i.e., so-called inter-sharing. In this paper, we provide an efficient distributed bandwidth management solution. This solution allows bandwidth sharing among backup paths of the same and different service LSPs, i.e., both intra-sharing and inter-sharing, with a guarantee of bandwidth protection for any single node/link failure. We also propose an efficient algorithm for backup path selection with the associated signaling extensions for additional information distribution and collection. To evaluate our schemes, we compare them via simulation with the basic MPLS fast reroute proposal, IETF RFC 4090, on two networks. Our simulation results show that using our bandwidth management scheme can significantly reduce restoration overbuild from about 250% to about 100%, and our optimized backup path selection can further reduce restoration overbuild to about 60%.     

Prediction-based routing as RWA in multilayer traffic engineering

Multilayer traffic engineering (MLTE) allows coping with ever-increasing and varying traffic demands in IP-over-Optical multilayer networks. It utilizes cross-layer TE (Traffic Engineering) techniques to provision optical lightpath capacity to the IP/MPLS (Internet Protocol/ Multi-Protocol Label Switching) logical topology on-demand. Such provisioning however causes optical connection arrival rates that pose strong performance requirements to Routing and Wavelength Assignment (RWA) strategies. Collecting up-to-date network information for the RWA with rapidly changing network states can be quite difficult. Exposing optical layer state information to the IP layer in the overlay model, or transforming this optical layer information in a workable representation in an integrated control plane is similarly problematic. Prediction-Based Routing (PBR) has been proposed as a RWA mechanism for optical transport networks; it bases routing not on possibly inaccurate or outdated network state, but instead on previous connections set-up. In this article, we propose to implement PBR as the RWA mechanism in the optical layer of a multilayer network, and use the predictive capabilities of PBR to expose dynamic optical network information into the multilayer traffic engineering algorithm with minimal control plane overhead. Some simulations show the benefits of using the PBR in the optical layer for MLTE purposes.     

QoS online routing and MPLS multilevel protection: a survey

A survey of MPLS protection methods and their utilization in combination with online routing methods is presented in this article. Usually, fault management methods pre-establish backup paths to recover traffic after a failure. In addition, MPLS allows the creation of different backup types, and hence MPLS is a suitable method to support traffic-engineered networks. In this article, an introduction of several label switch path backup types and their pros and cons are pointed out. The creation of an LSP involves a routing phase, which should include QoS aspects. In a similar way, to achieve a reliable network the LSP backups must also be routed by a QoS routing method. When LSP creation requests arrive one by one (a dynamic network scenario), online routing methods are applied. The relationship between MPLS fault management and QoS online routing methods is unavoidable, in particular during the creation of LSP backups. Both aspects are discussed in this article. Several ideas on how these actual technologies could be applied together are presented and compared.     

Demonstration of the highly reliable Hikari router network based on a newly developed disjoint path selection scheme

Integration of multiprotocol label switching functions and multiprotocol lambda switching functions can enhance the throughput of IP networks and remove bottlenecks that are derived from electrical packet processing. To enhance the packet forwarding capability, NTT proposed a photonic MPLS concept that includes MP/spl lambda/S, and demonstrated IP, MPLS, and photonic MPLS integrated router systems called the photonic MPLS router. This router system is now called the Hikari router. The word Hikari is Japanese meaning beam, light, lightwave, optical, photonic, and sunshine. The amount of IP data traffic has grown remarkably. Massive IP routers and flexible route control mechanisms are now required to cope with the increased amount of traffic. The Hikari router can offer two solutions utilizing photonic switching technologies, and photonic network operation and management technologies. The first solution is utilizing photonic switching technologies realized using optical-switch-based crossconnect systems. The other solution is realized using the MPLS and MP/spl lambda/S signaling protocol and photonic network protection functions. In this article we report on the implementation of the Hikari router systems, propose a newly developed disjoint path selection scheme for generalized MPLS networks with shared risk link group constraints, and demonstrate the signaling protocol and network protection functions. The demonstration system achieves a distributed optical path set-up/tear-down protocol with an extended constraint-based routing label distribution protocol. Fast self-healing through automatic protection switching and a new restoration scheme are also implemented. These functions are successfully implemented, and the performance is verified on a demonstration network. The protection switching scheme achieves protection in less than 20 ms, and the optical path restoration scheme achieves restoration in less than 500 ms.     

一种基于跳数和时延的MPLS自适应流量工程算法

提出了一种基于跳数和时延的MPLS自适应流量工程算法，该算法根据LSP的跳数和时延来进行流量分配，从而减少由传统路由算法而引起的网络拥塞，优化网络资源的利用。仿真结果表明，该算法简单易行，性能良好。     

QoS aware traffic grooming and integrated routing on IP over WDM networks

In this article, we consider traffic grooming and integrated routing in IP over WDM networks. The challenges of this problem come from jointly considering traffic grooming, IP routing, and lightpath routing and wavelength assignment (RWA). Due to the high bandwidth of optical fiber, there exists a mismatch between the capacity needed by an IP flow and that provided by a single lightpath. Traffic grooming is therefore used to increase the network utilization by aggregating multiple IP flows in a single lightpath. However, traffic grooming incurs additional delays that might violate Quality-of-Service (QoS) requirements of IP users. In this work, the tradeoff between traffic grooming and IP QoS routing is well-formulated as a mixed integer and linear optimization problem, in which the revenue from successfully provisioning IP paths is to be maximized. Problem constraints include IP QoS, routing, optical RWA, and the WDM network capacity. We propose a novel Lagrangean relaxation (LGR) algorithm to perform constraint relaxation and derive a set of subproblems. The Lagrangean multipliers are used in the proposed algorithm to obtain a solution in consideration of grooming advantage and resource constraints simultaneously. Through numerical experiments and comparisons between the proposed algorithm and a two-phase approach, LGR outperforms the two-phase approach under all experimental cases. In particular, the improvement ratio becomes even more significant when the ratio of IP flow to the wavelength capacity is smaller.     

Dynamic connection provisioning with shared protection in IP/WDM networks

Internet protocol (IP) traffic connections arrive dynamically at wavelength‐division multiplexing (WDM) network edges with low data rates compared with the wavelength capacity, availability, and quality‐of‐service (QoS) constraints. This paper introduces a scheme to be integrated into the control and management plane of IP/WDM networks to satisfy the availability and QoS required for IP traffic connections bundled onto a single wavelength (lightpath) in WDM networks protected by shared‐backup path protection (SBPP). This scheme consists of two main operations: (i) routing multi‐granular connections and traffic grooming policies, and (ii) providing appropriate shared protection on the basis of subscribers’ service‐level agreements in terms of data rate, availability, and blocking probability. Using the Markov chain process, a probabilistic approach is developed to conceive connection blocking probability models, which can quantify the blocking probability and service utilization of M:N and 1:N SBPP schemes. The proposed scheme and developed mathematical models have been evaluated in terms of bandwidth blocking ratio, availability satisfaction rate, network utilization, and connection blocking probability performance metrics. The obtained research results in this paper provide network operators an operational setting parameter, which controls the allocation of working and backup resources to dynamic IP traffic connections on the basis of their priority and data rate while satisfying their requirements in terms of bandwidth and availability. Copyright © 2013 John Wiley & Sons, Ltd.     

Internetworking connectionless and connection oriented networks

The use of connection-oriented (CO) networks for the transport of IP traffic is seen to have value to both users and service providers. Given the expectation that most endpoint-generated traffic will be in the form of connectionless (CL) IP datagrams, we address the problem of how to internetwork a CL (IP) network with a CO network. CO networks can be packet-switched or circuit-switched. Examples of packet-switched CO networks include ATM and MPLS networks, in which resource reservations are made at the ATM or shim layer, and IP-switch-based networks, in which resource reservations are made at the IP layer. Examples of circuit-switched networks include SONET/SDH and WDM networks that consist of programmable optical crossconnects. We consider the internetworking problem for two modes of operation of CO networks: provisioned, in which connections are set up a priori, and switched, in which connections are set up on demand. The main focus of this article is on the more complex problem: the internetworking of CL IP networks with CO networks operated in a switched mode. Our solution consists of (i) interworking user plane protocols with protocol conversion in some cases instead of always using protocol encapsulation, (ii) interworking routing protocols by either simply having gateways know routing information of both networks or having all nodes know routing information of both networks, and (iii) interworking signaling protocols by using application- or transport-layer end-to-end handshakes to trigger connection setups through the CO network. We demonstrate throughput improvements with our integrated routing interworking scheme over the MPOA IP-ATM internetworking solution for two example networks     

Lightpath routing for intelligent optical networks

An overview of current issues and challenges in lightpath routing for optical networks is given. An architecture is presented in which optical switches are deployed, usually in the core, to interconnect IP routers at the edges. Lightpath routing within this architecture follows the framework of generalized multiprotocol label switching. Our discussion pays particular attention to the aspects of optical routing that differ from routing in irrational IP networks. Such aspects include physical layer constraints, wavelength continuity, the decoupling of the control network topology from the data network topology, explicit routing with wavelength assignment, and diversity routing for fast protection. We also present an algorithmic framework for lightpath computation, highlighting the issue of wavelength continuity and the differences between lightpath computation and traditional IP route computation     

Strategy for Dynamic Routing and Grooming of Data Flows into Lightpaths in New Generation Network Based on the GMPLS Paradigm

This paper reports on a novel strategy and related algorithm for realizing dynamic routing and grooming into wavelengths of data flows (label switched paths, LSPs) in new generation optical networks based on generalized MPLS (GMPLS). The method allows arbitrary granularities of LSPs. The new generation network is modeled as a multi-layer network consisting of an IP/MPLS layer and an optical layer. In particular, the proposed solution adopts a dynamic routing algorithm based on the Dijkstra algorithm, that makes use of a weight system, integrated with a suitable method for grooming LSPs into wavelengths based on the packing criterion, thus harmonizing the features of MPLS packet flows whose bandwidth vary in a continuous range of values, with the optical world, where the wavelength bandwidth ranges according to discrete values. The weight system is based on the concepts of least resistance routing that allows to evenly distribute the traffic at the MPLS layer, while packing improves the use of optical resources by favoring more filled wavelengths with respect to the emptier ones. To assess the validity of the proposed solution a simulation model has been realized. The results obtained by simulation show that the packing criterion allows reducing the refused bandwidth from two down to about four times, for a network load of 70% and 55%, respectively, when compared with the alternative method named spreading. The dependence of the proposed solution on bandwidth granularity has been also investigated. Moreover, in order to demonstrate the superior performance of the proposed routing solution, a comparison between the proposed strategy with relevant solutions known in the literature, based on either a single or multi-layer approach, is also reported. In order to perform the comparison, all the reference routing solutions that have been considered adopt the packing method for LSP grooming into the lightpaths. The results show that our solution outperforms the others in terms of amount of traffic that can be on-line accommodated. For instance, assuming a blocking probability of 10^–3, the proposed solution is able to further reduce the refused bandwidth of the best routing algorithm considered in the analysis by a factor of three times, thanks to the knowledge of optical resource availability.     

三层动态网络联合优化选路算法

 IP/MPLS over WDM被认为是下一代互联网络的最理想解决方案,但是在目前以及以后的很长一段时间内,网络结构仍然是IP/MPLS over SDH over WDM这样三层网络共存的情况。为了能够有效降低三层动态网络中的业务阻塞率,本文提出了三层动态网络联合优化选路算法。该算法分为两种模式,在模式1中,上层网络中被阻塞的业务可以开放的利用下层网络的空闲资源;在模式2中,将三层网络整合成为一个网络,在这个综合网络中为每一个业务进行选路。为了能够实时的传输每一个到达的动态业务,在这两种模式中采用蚁群策略为动态业务进行选路,提出了一个新的前向蚂蚁路径选择概率模型以及路径质量评判参数,使得所选的路径能兼顾链路利用率和路径跳数。仿真实验表明,同其他算法相比,本文提出的算法可以有效地降低三层动态网络的阻塞率。