Influence of GMPLS Recovery Mechanisms on TCP Performance

Optical networks based on wavelength-division-multiplexing (WDM) techniques are very likely to be omnipresent in future telecommunication networks. Those networks are deployed in order to face the steady growth of traffic, which is for a large part Internet related. In the resulting IP-over-WDM scenario, TCP/IP constitutes an important fraction of the traffic transported over these networks. As IP networks are becoming increasingly mission-critical, it is of the utmost importance that these networks (and hence the supporting transport networks) be able to recover quickly from failures such as cable breaks or equipment outages. To that end, several IP-over-WDM network scenarios and corresponding protection and restoration strategies have been devised. It is clear that some trade-offs will have to be made in order to choose an appropriate strategy. In this paper, we investigate the effects of such recovery actions on the behavior of TCP, being the ubiquitous protocol used by today's network users. We examine the influence of different parameters such as the speed of recovery actions, changing length of the routes followed by the client data (TCP flows), changes in available bandwidth, etc. Thereby, we focus on what the TCP end-users care about, i.e., the number of bytes transported end-to-end within a certain time interval.     

Fault management in IP-over-WDM networks: WDM protection versus IPrestoration

We consider an IP-over-WDM network in which network nodes employ optical crossconnects and IP routers. Nodes are connected by fibers to form a mesh topology. Any two IP routers in this network can be connected together by an all-optical wavelength-division multiplexing (WDM) channel, called a lightpath, and the collection of lightpaths that are set up form a virtual topology. In this paper, we concentrate on single fiber failures, since they are the predominant form of failures in optical networks. Since each lightpath is expected to operate at a rate of few gigabits per second, a fiber failure can cause a significant loss of bandwidth and revenue. Thus, the network designer must provide a fault-management technique that combats fiber failures. We consider two fault-management techniques in an IP-over-WDM network: (1) provide protection at the WDM layer (i.e., set up a backup lightpath for every primary lightpath) or (2) provide restoration at the IP layer (i.e., overprovision the network so that after a fiber failure, the network should still be able to carry all the traffic it was carrying before the fiber failure). We formulate these fault-management problems mathematically, develop heuristics to find efficient solutions in typical networks, and analyze their characteristics (e.g., maximum guaranteed network capacity in the event of a fiber failure and the recovery time) relative to each other     

MPLS网络的自愈恢复技术

网络可靠性已经成为IP网的一个重要问题。目前关于IP网可靠性的讨论集中于多协议标记交换(MPLS)的故障恢复机制。在MPLS网络中的任何资源都有可能发生故障,要想提供一个高可用性的网络,网络提供商必须能够预料并且解决这些可能发生的故障。MPLS自愈恢复是MPLS流量工程的一种重要应用特性。自愈恢复能力对于提高MPLS网络的可用性和稳定性具有关键意义。本文首先介绍了MPLS网络的自愈恢复特性,并着重对目前ITU-T、ITEF推荐的几种MPLS故障检测、保护恢复技术进行深入介绍和对比分析。     

A novel approach to provision differentiated services in survivable IP-over-WDM networks

IP-over-WDM networks are starting to replace legacy telecommunications infrastructure and they form a promising solution for next-generation networks (NGNs). Survivability of an IP-over-WDM network is gaining increasing interest from both the Internet research community and service providers (SPs). We consider a novel static bandwidth-provisioning algorithm to support differentiated services in a survivable IP-over-WDM network. We propose and investigate the characteristics of both integer linear program (ILP) and heuristic approaches to solve this problem. In the heuristic method, we propose backup reprovisioning to ensure network resilience against single-node or multiple-link failures. Illustrative examples compare and evaluate the performance of the two methods in terms of capacity-usage efficiency and computation time.     

IP restoration vs. optical protection: Which one has the least bandwidth requirements?

Survivability in IP-over-WDM networks has already been extensively discussed in a series of studies. While many studies assume an IP restoration scheme and focus on network connectivity in order to ensure proper recovery, few studies deal with optical protection. We investigate this question with the objective of estimating the respective bandwidth requirements of both recovery schemes, subject to single or multiple failures. We also design a mixed recovery scenario where the recovery is taken care at a different layer depending on the type of failures.Results shows that optical protection is by far the most economical recovery scheme in terms of bandwidth requirements.     

Optimum Arrangement of Reliable Label-Switched Paths in MPLS Over Optical Networks

In MPLS (Multi-Protocol Label Switching) over optical networks, both the optical level and the MPLS level fault recovery can be considered. Generally, a more flexible path arrangement can be realized by the MPLS level recovery, while fast recovery can be achieved by the optical level recovery. When the optical level recovery is adopted, only normal traffic is carried through the working lightpaths and only recovered traffic is carried through the backup lightpaths. In contrast, the working LSPs (Label-Switched Paths) and the backup LSPs corresponding to other working LSPs can be accommodated into an identical lightpath when the MPLS level recovery is adopted. By such sophisticated accommodation of LSPs into the lightpaths, lightpath bandwidth can be utilized efficiently under the condition that the bandwidth utilization is restricted to attain the given objective of transfer quality for the MPLS packets in the normal state and unrestricted in a short time a failure occurs somewhere in the network. This paper proposes a simple mathematical programming model to obtain the optimum arrangement of the working and backup LSPs assuming the MPLS level recovery and a practical LSPs provisioning mode. By comparing the minimized network cost obtained from the optimum arrangement of the working and backup LSPs with the network cost resulting from the optical level recovery, this paper quantitatively evaluates the effectiveness of such bandwidth utilization improvement obtained from the MPLS level recovery and reveals that the MPLS level recovery can actually reduce the network cost due to its flexible arrangement of LSPs on the lightpaths.     

On the challenges of establishing disjoint QoS IP/MPLS paths across multiple domains

MPLS is being actively adopted as the core switching infrastructure at the intradomain level. This trend is mainly attributable to the undeniable potential of MPLS in terms of virtual private networks (VPNs) management, traffic engineering (TE), QoS delivery, path protection, and fast recovery from network failures. However, little progress has been made to attain the expected extension of MPLS label-switched paths (LSPs) across domain boundaries. Among the problems that remain unsolved is how to efficiently find and establish primary and protection interdomain LSPs for mission-critical services subject to QoS constraints. This article explores the major limitations hindering the deployment of these kinds of LSPs across multiple domains, in the context of the current interdomain network model. We describe the critical problems faced by the research community, and present our vision on how to rationally overcome some of the problems exposed. Our perspective is that we should be prepared for rather coarse-grained solutions as long as we need to coexist with the current interdomain network model     

Engineering end-to-end IP resilience usingresilience-differentiated QoS

Network resilience is becoming a key issue in the design of IP-based multimedia and multiservice networks. The current discussion about IP network resilience centers around MPLS-based recovery mechanisms. Any well designed recovery strategy has to take into account the different resilience requirements of the single traffic flows in order to avoid excessive usage of bandwidth for standby links. Faced with multiple recovery options, an ISP or NSP must decide which flows to protect to what extent against network failures. In this article an extension to existing quality of service (QoS) architectures is presented that integrates the signaling of resilience requirements with the traditional QoS signaling. We refer to this extended QoS model as resilience-differentiated QoS (RD-QoS). At the border of MPLS domains, the resilience requirements can then be directly mapped to the appropriate MPLS recovery options. A traffic engineering process for the provisioning of the resilience classes is introduced, and a case study demonstrates the significant network capacity savings achievable via this approach     

Broadband network restoration

In broadband networks, it is expected that at a given time there will be more data in the network than in narrowband networks, and therefore, to maintain the same quality of service, failures will need to be restored much faster. The purpose of this article is to demonstrate that broadband network technology, and in particular the asynchronous transfer mode (ATM), has factors that enable much faster restoration in broadband networks. These can be summarized as: (i) the higher-speed processing and larger-volume storage possible with today's technology as compared to what was available for narrowband networks, (ii) faster failure detection by making use of ATM cell header error checks, (iii) easier and faster rerouting simply by changing cell headers, (iv) easier and faster splitting of virtual paths by changing cell headers, (v) higher fill factors in the new digital hierarchy due to full termination at cross-connects and switches, and (vi) the statistical multiplexing advantage that can be exploited in a full ATM network configuration. The authors describe four basic characteristics of broadband network restoration, made faster and more efficient because of the factors listed above. These are (i) hitless protection switching, (ii) robust and fast failure detection, (iii) restoration alternatives in broadband networks, and (iv) fast topology update for multiple failures. By employing a combination of these methods, fast restoration of broadband networks can be achieved. The techniques described enable a B-ISDN/ATM network to be constructed that is highly self-healing (i.e., can recover from failures without human intervention) and very fast     

An efficient and flexible MPLS signaling framework for mobile networks

Multiprotocol Label Switching (MPLS) has gained momentum in recent years as an effective tool to provide Quality of Service (QoS) in a variety of networks. This has in turn created active interest in the area of recovery in MPLS based networks. A number of recovery schemes for MPLS domains have been proposed in recent years. However, the current schemes lack support for recovery in dynamic network topologies. In this paper, a new flexible signaling protocol for LSP rerouting in dynamic network environments is introduced. The signaling protocol recovers from node and link failures reactively, taking a local approach to LSP reestablishment. The performance of the signaling protocol is evaluated through simulations. Results indicate that the protocol can effectively and efficiently handle rerouting in dynamic networks with a low protocol signaling overhead as compared to contemporary MPLS rerouting protocols. This would enable the MPLS based IP-QoS support mechanisms to extend to dynamic network topologies. A preliminary version of this work was presented at the 2004 IEEE International Conference on Communications, Paris. Ramprasad Nagarajan has received his B.E. degree in Electronics and Telecommunications from Pune University, India in 1999. He received his M.S. degree in Electrical and Computer Engineering from the Ohio State University, Columbus, OH in 2004. Currently, he is a Wireless Network Engineer in Nortel Networks, specializing in the area of network architecture and design of wireless packet core networks. Ramprasad’s current research interests include the study of wireless network evolution trends, next generation wireless networks, network capacity planning, performance analysis, and optimization. He is a member of the IEEE. Eylem Ekici has received his B.S. and M.S. degrees in Computer Engineering from Bogazici University, Istanbul, Turkey, in 1997 and 1998, respectively. He received his Ph.D. degree in Electrical and Computer Engineering from Georgia Institute of Technology, Atlanta, GA, in 2002. Currently, he is an assistant professor in the Department of Electrical and Computer Engineering of the Ohio State University, Columbus, OH. Dr. Ekici’s current research interests include wireless sensor networks, vehicular communication systems, next generation wireless systems, and space-based networks, with a focus on routing and medium access control protocols, resource management, and analysis of network architectures and protocols. He also conducts research on interfacing of dissimilar networks.     

LDP failure detection and recovery

In the last few years, multiprotocol label switching has been successfully, deployed by the majority of service providers worldwide. The label distribution protocol is used in many MPLS networks for distributing labels to establish the label switched paths. This article focuses on LDP failures, namely failures that may occur in LDP while the underlining interior gateway protocol of choice, and the physical connections are operating normally. Since LDP and MPLS in general do not have intrinsic means for detecting these failures, new mechanisms have to be introduced to handle them. Furthermore, the IGP may remain unaware of the LDP failure, and continue to direct traffic to the faulty path. To resolve this situation, coupling between LDP and the routing protocol may have to be introduced. In this article, we discuss all aspects related to handling LDP failures: discovery, location, notification, recovery, and prevention.     

OAM mechanisms in MPLS layer 2 transport networks

This article describes OAM in MPLS layer 2 transport networks. MPLS networks used to transport layer 2 traffic are referred to as MPLS layer 2 transport networks. They may be used to connect legacy layer 2 networks and/or provide layer 2 service to a user over a MPLS network. As legacy layer 2 networks migrate to use MPLS for transport, the role of MPLS OAM mechanisms is becoming increasingly important. This is because the converged network must offer the same OAM functionality as existing layer 2 networks. This article emphasizes the importance of end-to-end OAM, while emulating existing layer 2 services using MPLS transport. End-to-end fault detection is described in the context of various layer 2 over MPLS transport network models. The article focuses on state-of-the-art MPLS label switched path and pseudo wire OAM mechanisms being developed by the IETF. This includes fault detection and isolation mechanisms such as LSP-Ping, bidirectional forwarding detection, and virtual circuit connectivity verification. The applicability of each of these mechanisms is given. In some cases it may be possible to carry layer 2 OAM cells end-to-end, while in other cases this may not be possible. The relationship between segment-based OAM mechanisms and end-to-end OAM is described for each of these cases.     

一种基于RSVP-TE的MPLS网络FRR机制

快速重路由FRR用于MPLS网络故障时的修复,能够降低故障导致的传输时延和丢包率,保障业务的服务质量,满足网络中有特殊需求的应用。介绍了基于RSVP-TE协议的FRR原理,重点阐述了通过RSVP-TE交互,在MPLS网络中对重要链路或节点预先建立备份路径,检测到链路或节点故障后进行修复的解决方案,给出了具体实现框架,介绍了备份路径的建立过程以及检测故障后的处理过程。介绍了该机制在网络中的典型应用和前景。     

Recovery in multilayer optical networks

The integration of different network technologies into a multilayer network, as in Internet-based networks carried by optical transport networks (OTNs), creates new opportunities but also challenges with respect to network survivability. In different network layers, recovery mechanisms that are active can be exploited jointly to reach a more efficient or faster recovery from failures. This interworking is also indispensable in order to overcome the variety of failure scenarios that can occur in the multilayer-network environment. A well-considered coordination between the different layers and their recovery mechanisms is crucial in order to attain high performance recovery. This paper provides an overview of multilayer recovery issues and solutions in an Internet protocol (IP)-over-optical-network environment, which is illustrated by quantitative case studies.     

A comparison of linear and mesh topologies-DQDB and the Manhattanstreet network

The authors make a performance and reliability comparison of two networks that have been proposed for use as high-speed metropolitan area networks (MANs)-the Manhattan street network (MSNet) and the distributed queue dual bus (DQDB) network. Both networks use slotted access protocols and have the same number of links, transmitters, and receivers per node. The DQDB network has been adopted by the IEEE 802.6 committee as the metropolitan area network (MAN) standard. The authors demonstrate the relative superiority of the MSNet over the DQDB network. They show that the MSNet provides a much higher network throughput for a variety of traffic patterns-uniform and nonuniform. They also look at the reliability of both networks and show that the MSNet can survive more failures than the DQDB network and that failures cause a lesser performance degradation in the MSNet. It is also shown that higher-level mechanisms are required in the DQDB network to recover from link failures whereas deflection routing is sufficient in the MSNet     

Dynamic recovery for survivable virtual network embedding

Network virtualization is a promising way to overcome the current ossification of the Intemet. It is essential challenge to find effective, efficient and robust embedding algorithms for recovering virtual network. The virtual network mapping algorithm based on integer programming which was proposed months ago. But it did consider the faults of physical network resources, which is so called survivable virtual network embedding （VNE） problem. Previous strategies for enabling survivability in network virtualization focused on providing protection for the physical network or enhancing the virtual networks by providing backup physical resources in advance, and treated all the physical failures as link failures. In the article, a dynamic recovery method is proposed to solve the survivable virtual network embedding problem based on the integer programming VNE algorithm. The dynamic recovery method doesn＇t need to backup physical resources and it makes more substrate resources which can be used in the embedding. The dynamic recovery process will be activated only when physical failures occur. Different algorithms are used to recovery node and link failures. Simulations show that the method helps to recover almost all of physical failures by finding the substitute nodes and paths, and its performance is very close to that of pure VNE method without considering physical failures.     

Hybrid FRR/p-cycle design for link and node protection in MPLS networks

Survivable MPLS technologies are crucial in ensuring reliable communication services. The fast reroute (FRR) mechanism has been standardized to achieve fast local repair of label switched paths (LSPs) in the event of link or node failures. We present a suite of hybrid protection schemes for MPLS networks that combine the well-known p-cycle method with FRR technology. Whereas with pure FRR backup paths are planned by each node individually, the hybrid schemes employ a set of p-cycles that may be selected using techniques that take a holistic view of the network so as to share protection bandwidth effectively. The hybrid FRR/p-cycle methods are fully RFC 4090-compliant, yet allow network operators to leverage a large existing body of p-cycle design techniques. Numerical results on realistic network topologies indicate that the hybrid approach is successful in combining the advantages of p-cycle design and FRR.     

Differentiated QoS for survivable WDM optical networks

Optical networks based on WDM technology have become a promising solution to realize transport networks that can meet the ever-increasing demand for bandwidth. As WDM networks carry a huge volume of traffic, maintaining a high level of survivability is an important and critical issue. The. development of GMPLS switching technology led to the direct integration of IP and WDM. In these IP-over-WDM networks different applications/end users need different levels of fault tolerance and differ in how much they are willing to pay for the service they get. The current trend in network development is moving toward a unified solution providing support for voice, data, and various multimedia services. Therefore, it imperative that WDM networks incorporate fault tolerance to single or multiple component failures, protection bandwidth, recovery time, and recovery granularity besides resource utilization and call acceptance ratio. This article presents a survey of various methods that have been proposed for providing service differentiation in survivable WDM networks and discuss their performance. Such methods are broadly classified under various paradigms such as differentiated reliability, R-connections, quality of protection, and quality of recovery.     

Survivable routing in IP-over-WDM networks: An efficient and scalable local search algorithm

In IP-over-WDM networks, a logical IP network is routed on top of a physical optical fiber network. An important challenge here is to make the routing survivable. We call a routing survivable if the connectivity of the logical network is guaranteed in the case of a failure in the physical network. In this paper we describe FastSurv, a local search algorithm for survivable routing. The algorithm works in an iterative manner: after each iteration it learns more about the structure of the logical graph and in the next iteration it uses this information to improve its solution. The algorithm can take link capacity constraints into account and can be extended to deal with multiple simultaneous link failures and node failures. In a large series of tests we compare FastSurv with current state-of-the-art algorithms for this problem. We show that it can provide better solutions in much shorter time, and that it is more scalable with respect to the number of nodes, both in terms of solution quality and run time.     

Dual-Link Failure Resiliency Through Backup Link Mutual Exclusion

Networks employ link protection to achieve fast recovery from link failures. While the first link failure can be protected using link protection, there are several alternatives for protecting against the second failure. This paper formally classifies the approaches to dual-link failure resiliency. One of the strategies to recover from dual-link failures is to employ link protection for the two failed links independently, which requires that two links may not use each other in their backup paths if they may fail simultaneously. Such a requirement is referred to as backup link mutual exclusion (BLME) constraint and the problem of identifying a backup path for every link that satisfies the above requirement is referred to as the BLME problem. This paper develops the necessary theory to establish the sufficient conditions for existence of a solution to the BLME problem. Solution methodologies for the BLME problem is developed using two approaches by: 1) formulating the backup path selection as an integer linear program; 2)developing a polynomial time heuristic based on minimum cost path routing. The ILP formulation and heuristic are applied to six networks and their performance is compared with approaches that assume precise knowledge of dual-link failure. It is observed that a solution exists for all of the six networks considered. The heuristic approach is shown to obtain feasible solutions that are resilient to most dual-link failures, although the backup path lengths may be significantly higher than optimal. In addition, the paper illustrates the significance of the knowledge of failure location by illustrating that network with higher connectivity may require lesser capacity than one with a lower connectivity to recover from dual-link failures.