Showing papers in "Journal of Optical Networking in 2006"

Methodology and input availability parameters for calculating OpEx and CapEx costs for realistic network scenarios

Abstract: Feature Issue on High Availability in Optical NetworksThe availability requirements for today's networks are very high. Higher availability often comes with a higher cost. We describe several steps required for estimating the costs of realistic network scenarios. Capital expenditures (CapEx) and operational expenditures (OpEx) are classified. An activity-based approach is used to quantify the cost of the event-driven operational processes such as repair and service provisioning. We discuss activity duration and availability parameters as required input data, which are necessary for calculating the processes' costs for realistic network scenarios. The relevant availability measures for an IP-over-Optical network are described using a triplet representation with optimistic, nominal, and conservative values. The model is applied to a reference German network scenario.

Bandwidth management for WDM EPONs

Abstract: We explore the problem of bandwidth management for the evolutionary upgrade of WDM EPONs. We divide the bandwidth management problem into two subproblems: (1) grant sizing and (2) grant scheduling. We then apply a scheduling theoretical approach to find a best scheduler for WDM EPONs. We show by means of extensive simulations that a multidimensional scheduling approach using results from scheduling theory can provide much better bandwidth management by means of better wavelength utilization than a static wavelength assignment. We also show that an online scheduling approach can provide lower queueing delays than a cyclical offline scheduling approach. We conclude with some specific guidance on future research on bandwidth management for WDM EPONs.

Framework for waveband switching in multigranular optical networks: part I-multigranular cross-connect architectures [Invited]

Abstract: Feature Issue on Waveband Switching, Routing, and GroomingOptical networks using wavelength-division multiplexing (WDM) are the foremost solution to the ever-increasing traffic in the Internet backbone. Rapid advances in WDM technology will enable each fiber to carry hundreds or even a thousand wavelengths (using dense-WDM, or DWDM, and ultra-DWDM) of traffic. This, coupled with worldwide fiber deployment, will bring about a tremendous increase in the size of the optical cross-connects, i.e., the number of ports of the wavelength switching elements. Waveband switching (WBS), wherein wavelengths are grouped into bands and switched as a single entity, can reduce the cost and control complexity of switching nodes by minimizing the port count. This paper presents a detailed study on recent advances and open research issues in WBS networks. In this study, we investigate in detail the architecture for various WBS cross-connects and compare them in terms of the number of ports and complexity and also in terms of how flexible they are in adjusting to dynamic traffic. We outline various techniques for grouping wavelengths into bands for the purpose of WBS and show how traditional wavelength routing is different from waveband routing and why techniques developed for wavelength-routed networks (WRNs) cannot be simply applied to WBS networks. We also outline how traffic grooming of subwavelength traffic can be done in WBS networks. In part II of this study [Cao , submitted to J. Opt. Netw.], we study the effect of wavelength conversion on the performance of WBS networks with reconfigurable MG-OXCs. We present an algorithm for waveband grouping in wavelength-convertible networks and evaluate its performance. We also investigate issues related to survivability in WBS networks and show how waveband and wavelength conversion can be used to recover from failures in WBS networks.

Dynamic routing for shared path protection in multidomain optical mesh networks

Abstract: The routing problem for shared path protection in multidomain optical mesh networks is more difficult than that in single-domain mesh networks due to the lack of complete and global knowledge of the network topology and bandwidth allocation. To overcome this difficulty, we propose an aggregated network modeling by underestimation with a two-step routing strategy. In the first step, a rough routing solution is sketched in a virtual network that is the topology aggregation of the multidomain network. A complete routing is then determined by solving routing problems within the original single-domain networks. The first step can be solved by either using an exact mathematical program or a heuristic, whereas the second step is always solved by heuristics. Computational results show the relevance of the aggregated network modeling. They also prove the scalability of the proposed routing for multidomain networks and its efficiency in comparison with the optimal solution obtained by use of the complete information scenario. In addition, we believe that short working paths lead to a higher possibility of sharing backup resources between backup paths. Our mathematical program model minimizes the total requested resources and at the same time provides a short working path, resulting in a further overall saving of resources.

EPON versus APON and GPON: a detailed performance comparison

Abstract: Feature Issue on Optical Ethernet (OE) Ethernet passive optical network (EPON) efficiency issues in both upstream and downstream directions are discussed in detail, describing each component of the overall transmission overhead as well as quantifying their effect on the system's performance and comparing them with the other existing passive optical network (PON) access systems, namely, asynchronous transfer mode PON (APON) and generic framing PON (GPON). For EPON, two main transmission overhead groups are defined, namely, Ethernet encapsulation overhead and EPON-specific scheduling overhead. Simulations are performed using the source aggregation algorithm (SAA) to verify the Ethernet encapsulation overhead for various synthetic and measured packet size distributions (PSDs). A SAA based an EPON simulator is used to verify both upstream and downstream overall channel efficiencies. The obtained simulation results closely match the theoretical limits estimated based on the IEEE 802.3ah standard. An estimated throughput of 820 to 900 Mbits/s is available in the upstream direction, whereas in the downstream direction effective throughput ranges from 915 to 935 Mbits/s.

Selective randomized load balancing and mesh networks with changing demands

Abstract: We consider the problem of building cost-effective networks that are robust to dynamic changes in demand patterns. We compare several architectures using demand-oblivious routing strategies. Traditional approaches include single-hop architectures based on a (static or dynamic) circuit-switched core infrastructure and multihop (packet-switched) architectures based on point-to-point circuits in the core. To address demand uncertainty, we seek minimum cost networks that can carry the class of hose demand matrices. Apart from shortest-path routing, Valiant's randomized load balancing (RLB), and virtual private network (VPN) tree routing, we propose a third, highly attractive approach: selective randomized load balancing (SRLB). This is a blend of dual-hop hub routing and randomized load balancing that combines the advantages of both architectures in terms of network cost, delay, and delay jitter. In particular, we give empirical analyses for the cost (in terms of transport and switching equipment) for the discussed architectures, based on three representative carrier networks. Of these three networks, SRLB maintains the resilience properties of RLB while achieving significant cost reduction over all other architectures, including RLB and multihop Internet protocol/multiprotocol label switching (IP/MPLS) networks using VPN-tree routing.

Path-protection-based routing and wavelength assignment in wavelength-division multiplexing optical networks under a scheduled traffic model

Abstract: Feature Issue on High Availability in Optical NetworksWe study the optimal survivability design problem under a scheduled traffic model in wavelength-convertible WDM optical mesh networks. In this model, a set of demands is given, and the setup and tear-down time of a demand are known in advance. We formulate the joint routing and wavelength assignment problems as integer linear problems that maximally exploit network resource reuse in both space and time. Moreover, we propose efficient heuristic algorithms with different demand-ordering policies to solve large survivable routing and wavelength assignment problems under the scheduled traffic model. Our simulation results indicate that the optimization of resource sharing in space and time enabled by our connection-holding-time-aware protection schemes can achieve significantly better resource utilization than schemes that are holding-time unaware. In addition, the proposed heuristic algorithms are shown to be indeed very effective.

Ring-based local access PON architecture for supporting private networking capability

Abstract: We propose what we believe is a novel ring-based local access Passive Optical Network (PON) architecture that addresses some of the limitations of current tree-based PON architectures including supporting private networking capability Specifically, we propose and devise a simple ring-based Ethernet PON (EPON) architecture that supports a truly shared LAN capability among end users as well as upstream access to the central office Unlike a typical ring-based PON topology in which the optical line terminal (OLT) and the optical network units (ONUs) are interconnected via a long fiber ring, under the proposed architecture, ONUs are interconnected via a short distribution fiber ring in the local loop but share the standard trunk feeder fiber for long reach connectivity to the Central Office (CO) The main characteristic of the proposed architecture is that it supports a fully distributed control plane among the ONUs for ONU-ONU communication as well as upstream access to the OLT This architecture is well suited for an autonomous access environment such as a university campus or a private corporation where several buildings are closely dispersed within a 05-1 km diameter area

IRIS optical packet router [Invited]

Abstract: Feature Issue on ConvergenceWe present a load-balanced packet router with an all-optical data plane and a decentralized control plane. The router, whose architecture scales up to 256 Tbit/s with current technology, consists of two space switches based on wavelength converters and large N×N arrayed-waveguide gratings, surrounding a deterministic time buffer. First experimental results have been obtained in a 2×2 testbed with 40 Gbit/s wavelength converters that use monolithically-integrated semiconductor optical amplifiers, and with fiber-delay-line time buffers.

How much wavelength conversion allows a reduction in the blocking rate

Abstract: We study the problem of routing and wavelength assignment (RWA) in a WDM optical network under different hop assumptions, i.e., with and without wavelength converters, considering the objective of minimizing the blocking rate. We design a heuristic with two interactive phases, one for the routing and one for the wavelength assignment, which generalizes a previous algorithm by Noronha and Ribeiro [Eur. J. Oper. Res. 171, 797 (2006)] based on a Tabu Search scheme using a partition coloring reformulation for uniform traffic and single-hop connections. Considering nonuniform traffic, we explore a reformulation of the RWA problem as a generalized partition coloring problem and develop a Tabu Search algorithm to solve it. We also explore how to integrate multihop connections, with the addition of conversion features at some or at all optical nodes. Experiments are done on several traffic and network instances. Most heuristic solutions are excellent as illustrated by the very small gap between the values provided by the heuristic and the optimal values of the linear relaxation. We next show that conversion features, although often considered an added value, are of little help in improving on the blocking rate except for some very particular traffic instances, even on realistic network topologies.

Self-protection scheme against failures of distributed fiber links in an Ethernet passive optical network

Abstract: A novel self-protection scheme for an Ethernet passive optical network is introduced and studied at both the physical and the media access control layers. The scheme is simple and fast and can provide 1:1 protection and automatic traffic restoration against the fiber link failure between a remote node (RN) and any optical network unit (ONU). Simulation results show that fiber failure does not degrade the transmission performance, and the restoration time depends mainly on the switch time of the physical layer. Our protection scheme saves many long fibers, does not influence other normal ONUs, and requires no active device in the RN.

Protection in multigranular waveband networks

Abstract: Feature Issue on Waveband Switching, Routing, and Grooming As the number of wavelengths in wavelength-division-multiplexed (WDM) systems continues to increase, the switching fabrics of optical cross connects (OXCs) will become increasingly complex. This complexity can be reduced by introducing multigranular optical cross connects (MG-OXCs) into the network. A MG-OXC is capable of switching an entire group of wavelengths through a single switch port, thereby reducing the required number of switching ports and the complexity of the switch fabric. We consider the problem of establishing active and backup paths for connections in networks with MG-OXCs. The problem of establishing protected connections for a given traffic demand in multigranular networks differs from the protection problem in standard wavelength-routed networks in that the primary objective in multigranular networks is to minimize the number of switch ports in the network. The mode of protection (dedicated or shared) affects the ability to aggregate and route traffic together. We formulate the problems of dedicated and shared protection in the form of Integer Linear Programs (ILPs) to optimize the total number of ports. Since solving the ILPs for large networks is intractable, we propose a graph-based heuristic that attempts to solve the problem of routing and waveband assignment in an integrated manner. The heuristic reduces the total number of ports in the multigranular network. We also study the effect of shared protection on the port count.

Algorithm for protecting light-trees in survivable mesh wavelength-division-multiplexing networks

Abstract: Wavelength-division-multiplexing (WDM) technology is expected to facilitate bandwidth-intensive multicast applications such as high-definition television. A single fiber cut in a WDM mesh network, however, can disrupt the dissemination of information to several destinations on a light-tree based multicast session. Thus it is imperative to protect multicast sessions by reserving redundant resources. We propose a novel and efficient algorithm for protecting light-trees in survivable WDM mesh networks. The algorithm is called segment-based protection with sister node first (SSNF), whose basic idea is to protect a light-tree using a set of backup segments with a higher priority to protect the segments from a branch point to its children (sister nodes). The SSNF algorithm differs from the segment protection scheme proposed in the literature in how the segments are identified and protected. Our objective is to minimize the network resources used for protecting each primary light-tree such that the blocking probability can be minimized. To verify the effectiveness of the SSNF algorithm, we conduct extensive simulation experiments. The simulation results demonstrate that the SSNF algorithm outperforms existing algorithms for the same problem.

Capacity versus availability trade-offs for availability-based routing

Abstract: Feature Issue on High Availability in Optical NetworksThe core idea of availability-based routing consists of exploiting the availability as a link-state metric: this kind of approach has received attention by virtue of the ability to avoid excessively unavailable paths. A recent availability-based heuristic design technique for reliable optical transport networks [IEEE Sel. Areas Commun. 23, 1520 (2005)] showed that the availability metric returns results in line with the traditional length or hop metric, due to the close relationship between availability and length of the link. We break the direct relationship between length and availability by considering a highly available backbone next to a less reliable peripheral section of the network. Under this availability-unbalanced scenario, we analyze the trade-offs between capacity and availability: pros and cons of the availability metric emerge much more clearly. Performance of an availability-based metric is shown and discussed applying dedicated path protection. In addition we provide, for the first time to our knowledge, analytical proof of the optimality (under certain hypothesis) of the proposed maximal availability routing approach for dedicated path protection.

Architecture design and performance evaluation of multigranularity optical networks based on optical code division multiplexing

Abstract: Feature Issue on Waveband Switching, Routing, and GroomingIn traditional lambda-based multigranularity optical networks, a lambda is always treated as the basic routing unit, resulting in low wavelength utilization. On the basis of optical code division multiplexing (OCDM) technology, a novel OCDM-based multigranularity optical cross-connect (MG-OXC) is proposed. Compared with the traditional lambda-based MG-OXC, its switching capability has been extended to support fiber switching, waveband switching, lambda switching, and OCDM switching. In a network composed of OCDM-based MG-OXCs, a single wavelength can be shared by distinct label switched paths (LSPs) called OCDM-LSPs, and OCDM-LSP switching can be implemented in the optical domain. To improve the network flexibility for an OCDM-LSP provisioning, two kinds of switches enabling hybrid optical code (OC)-wavelength conversion are designed. Simulation results indicate that a blocking probability reduction of 2 orders can be obtained by deploying only five OCs to a single wavelength. Furthermore, compared with time-division-multiplexing LSP (TDM-LSP), owing to the asynchronous accessibility and the OC conversion, OCDM-LSPs have been shown to permit a simpler switch architecture and achieve better blocking performance than TDM-LSPs.

Resilience design in all-optical ultralong-haul networks

Abstract: Feature Issue on High Availability in Optical NetworksAfter many years of research and industry efforts, ultralong-haul (ULH) technologies for DWDM transport are maturing, and carriers are deploying them for high capacity and capital savings. Unlike opaque optical networks, it is still uncertain how best to restore an all-optical ULH network after failure. This paper focuses on this issue and proposes a standby restoration scheme and its variants with different restoration speed and associated restoration cost. Our observations suggest that the proposed hot standby scheme would yield both faster restoration and cost efficiencies compared with other schemes.

Prioritized retransmission in slotted all-optical packet-switched networks

Abstract: Feature Issue on Photonics in SwitchingWe consider an all-optical slotted packet-switched network interconnected by a number of bufferless all-optical switches with contention-based operation. One approach to reduce the cost of the expensive contention resolution hardware could be retransmission in which each ingress switch keeps a copy of the transmitted traffic in the electronic buffer and retransmits whenever required. The conventional retransmission technique may need a higher number of retransmissions until traffic passes through the network. This in turn may lead to a retransmission at a higher layer and reduce the network throughput. In this paper, we propose and analyze a simple but effective prioritized retransmission technique in which dropped traffic is prioritized when retransmitted from ingress switches so that the core switch can process them with a higher priority. We present the analysis of both techniques in multifiber network architecture and verify it via simulation to demonstrate that our proposed algorithm can limit the number of retransmissions significantly and can improve TCP throughput better than the conventional retransmission technique.

Optical network architectures for dynamic reconfiguration of full duplex,multiwavelength, radio over fiber

Abstract: In recent years there have been rapid advances in the techniques to generate and transport radio signals over optical fiber, but very little has been done to date to develop the concept of a full duplex, multiwavelength, radio-over-fiber network architecture that intrinsically supports and facilitates the dynamic reconfiguration of the wireless network. Recently the development of suitable architectures has received attention, with some approaches concentrating on the fiber radio link layer [Fiber Integr. Opt.19, 167 (2000)] and others focusing on the optical wavelength division multiplexing layer [L. Lim , in International Topical Meeting on Microwave Photonics: MWP'99 (IEEE, 1999), p. 127; also Fiber Opt. Business (15 November 2000)]. Here we present, for the first time to our knowledge, an approach that aims to vertically integrate the cellular radio layer, the fiber radio layer, the optical networking layer, and the physical layer to achieve a network architecture that enables the dynamic reconfiguration of the cellular wireless network layer.

Hierarchical interdomain routing and light-path provisioning in optical networks

Abstract: Given the increasing adoption of DWDM networking technology, there is now a growing need to address distributed interdomain light-path provisioning issues. Although interdomain provisioning has been well studied for packet/cell-switching networks, the wavelength dimension presents many additional challenges. To address this challenge, a detailed hierarchical routing GMPLS-based framework for provisioning all-optical and optoelectronic multidomain DWDM networks is presented. The scheme adapts topology abstraction schemes to improve routing scalability and lower interdomain blocking probabilities. Related interdomain light-path RWA and signaling schemes are also tabled. Performance analysis results are also presented to demonstrate the effectiveness of the proposed mechanisms along with directions for future research work.

Design of wavelength-convertible edge nodes in wavelength-routed networks

Abstract: Wavelength converters reduce the connection blocking probability in wavelength-routed networks by eliminating the wavelength continuity constraint. We develop a method for deployment of wavelength converters in wavelength-routed networks with an overlay model. In these networks, most wavelength converters are deployed on edge nodes to cover the difference in the numbers of wavelengths multiplexed on access and core links. Therefore reduction of wavelength converter cost on edge nodes leads to minimizing the wavelength converter cost in the whole network. We propose an ingress edge node architecture with fixed wavelength converters that have limited wavelength convertibility but are more economical than full wavelength converters. In our architecture, each input access link of ingress edge nodes is equipped with fixed wavelength converters, and input wavelengths from the access links are evenly distributed on the output core link. As a result, competition for a free wavelength on an output core link is avoided. Simulation results show that our edge node architecture offers about 20% cost reduction compared with a node architecture that uses only full wavelength converters where networks are actually under operation and a full wavelength converter cost to fixed wavelength converter cost ratio is 3:1.

Heterogeneous waveband switching in wavelength division multiplexed networks based on autonomous clustering architecture [Invited]

Abstract: Feature Issue on Waveband Switching, Routing, and GroomingAdopting waveband switching (WBS) in backbone wavelength division multiplexed (WDM) networks is promising since it can reduce the network operational cost and the call blocking probability. However, upgrading the existing optical switching architecture requires time and money. Thus heterogeneous waveband switching (HeteroWBS) architecture would be desirable in such a system, where some nodes can support WBS functions and some cannot. We study the performance of HeteroWBS networks in terms of call blocking probability and cost savings under dynamically arriving traffic requests. We first investigate the effects of optical component developments on waveband switching in WDM networks. Various connection managements are then listed and analyzed. Next, to assist in the designing of efficient WBS algorithms, an autonomous clustering-based HeteroWBS (AS-HeteroWBS) architecture is proposed. The AS-HeteroWBS architecture clusters the network into multiple autonomous systems (ASs). An AS may contain some specific nodes that provide WBS functions for all the nodes in the AS. Based on the architecture, three HeteroWBS algorithms are proposed, namely, the autonomous heterogeneous WBS algorithm (AS-WBS), the autonomous source-limited heterogeneous WBS algorithm (AS-S-WBS), and the shortest-path-based heterogeneous WBS algorithm (SH-WBS). Our simulation results show that the HeteroWBS algorithms can achieve optimal cost savings while maintaining the same network throughput compared with the algorithm without WBS.

Scalable, distributed cycle-breaking algorithms for gigabit Ethernet backbones

Abstract: "http://www.osa-jon.org/virtual_issue.cfm?vid=7" Feature Issue on Optical Ethernet (OE)Ethernet networks rely on the so-called spanning-tree protocol (IEEE 802.1d) to break cycles, thereby avoiding the possibility of infinitely circulating packets and deadlocks. This protocol imposes a severe penalty on the performance and scalability of large gigabit Ethernet backbones since it makes inefficient use of fibers and may lead to bottlenecks. We propose a significantly more scalable cycle-breaking approach based on the theory of turn prohibition. Specifically, we introduce, analyze, and evaluate an algorithm called tree-based turn prohibition (TBTP). We show that this polynomial-time algorithm maintains backward compatibility with the IEEE 802.1d standard and never prohibits more than half the turns in the network for any given graph and any given spanning tree. We further introduce a distributed version of the algorithm that nodes in the network can run asynchronously. Through extensive simulations on a variety of graph topologies, we show that the TBTP algorithm can lead to an order of magnitude improvement over the spanning-tree protocol with respect to throughput and end-of-end delay metrics. In addition, we propose and evaluate heuristics to determine the replacement order of legacy switches that results in the fastest performance improvement.

Complete rerouting protection

Abstract: Feature Issue on AvailabilityProtection of communication against network failures is becoming increasingly important, and here we present what we believe to be the most capacity-efficient protection method possible, the complete rerouting protection method, when we require that all communication should be restored in the case of a single link network failure. We present a linear programming model of the protection method and a column-generation algorithm. For six real-world networks, the minimal restoration overbuild network capacity is between 13% and 78%. We further study the importance of the density of the network, derive analytical bounds, and study methods to speed up the column-generation algorithm. We expect that the suggested protection method will be used for calculating lower bounds required for protection, but not as a functioning protection method.

Single-layer multigranular optical cross-connect architecture with conversion capability and enhanced flexibility

Abstract: Feature Issue on Waveband Switching, Routing, and GroomingThe rapid advances in WDM technology are expected to bring about tremendous growth in the size of optical cross connects (OXCs). In this context, multigranular OXCs (MG-OXCs) have been suggested as a means of reducing the amount of equipment required. Here we expand the concept of MG-OXCs to include optical packet granularity and review the key building blocks for the advent of MG-OXCs. A single-layer MG-OXC is suggested that offers enhanced flexibility with respect to other single-layer concepts, conversion capability, and good physical performance. Concatenation performance is analytically investigated.

Segment protection algorithm based on an auxiliary graph for wavelength-division multiplexing optical networks

Abstract: Shared-segment protection is a promising protection scheme in wavelength-division multiplexing optical networks. In segment protection, there exist multiple strategies to partition a given working path, and different strategies lead to different network performance. According to the strategy used to partition a working path, we classify segment protection into predetermined partitioning, postdetermined partitioning, and integrated partitioning. We show that integrated partitioning combines the advantages of the other two strategies. In addition, we propose a new heuristic using an integrated partitioning strategy, called an auxiliary-graph-based segment protection (AGBSP) algorithm. Based on an auxiliary graph, AGBSP dynamically selects an optimal set of backup segments from all predetermined possible backup segments for a given working path. We compare AGBSP with shared-path protection and other efficient segment protection algorithms proposed in the literature. The results show that AGBSP achieves the best performance in terms of blocking probability and resource overbuild ratio, and it has a short recovery time.

Wavelength assignment scheme for WDM networks with limited-range wavelength converters

Abstract: We propose a new wavelength assignment scheme that improves the blocking probability of WDM networks that use limited-range wavelength converters. Limited-range wavelength converters are attractive for wavelength-routed networks, given current technology, since they offer good utilization of the wavelength resource and improved blocking probability. However, their conversion ranges are limited; the maximum difference between the input and the output wavelengths is restricted. Thus we must take into account the existence of these limited-range wavelength converters. In our proposed scheme, each connection request is assigned a different wavelength according to its hop number. We tend to use different wavelengths for connection requests with different hop numbers. As a result, we can reduce the blocking probability by two decades compared with simply assigning the shortest available wavelengths. In addition, the scheme allows the number of wavelength converters used in each node to be reduced with almost no degradation in blocking probability. Simulation results show that the proposed scheme can reduce the wavelength converters by about 20% on the simple ring network and by 37.5% on the 14-node NSFNet network.

Dynamic segment shared protection for multicast traffic in meshed wavelength-division-multiplexing optical networks

Abstract: We investigate the protection approach for dynamic multicast traffic under shared risk link group (SRLG) constraints in meshed wavelength-division-multiplexing optical networks We present a shared protection algorithm called dynamic segment shared protection for multicast traffic (DSSPM), which can dynamically adjust the link cost according to the current network state and can establish a primary light-tree as well as corresponding SRLG-disjoint backup segments for a dependable multicast connection A backup segment can efficiently share the wavelength capacity of its working tree and the common resources of other backup segments based on SRLG-disjoint constraints The simulation results show that DSSPM not only can protect the multicast sessions against a single-SRLG breakdown, but can make better use of the wavelength resources and also lower the network blocking probability

Reliable services with fast protection in IP/MPLS over optical networks

Abstract: Feature Issue on High Availability in Optical NetworksWe define new quality-of-service (QoS) routing schemes with protection in Internet Protocol (IP)/Multiprotocol Label Switching (MPLS) over optical networks. The novelty of the proposed routing schemes is the use of the knowledge of the logical links already protected by the optical layer. The logical topology defined by the optical layer is given and fixed, and we assume that it is partially protected. Thereby, at the IP/MPLS layer, spare capacity is reserved to protect only those links that are unprotected. Moreover, we also characterize the traffic services based on their level of reliability and QoS requirements. In order to guarantee fast protection, segment protection and shared backups are combined, resulting in suitable fault recovery time and resource consumption. A complete set of experiments proves that the proposed schemes are more efficient than the previous ones in terms of resources used to protect the network, failure impact, and blocking probability.

Convergence through packet-aware transport [Invited]

Abstract: Feature Issue on ConvergenceWe provide a comprehensive description of the Packet-Aware Transport Network (PATN) architecture first introduced in [IEEE Commun. Mag. (3) 120 (2004)]. The PATN architecture provides a more efficient transport of data services such as Ethernet and IP than traditional time-division multiplexing- (TDM) centric access network architectures. The PATN architecture builds upon several emerging technologies and improvements in silicon processing to seamlessly converge multiple network functions onto a single platform, optimize network utilization, and reduce operating costs. The PATN architecture is also designed to support all existing TDM voice, private line, and data services. To validate this architecture we implemented a comprehensive test bed that provides a microcosm of local, metro, and long-distance transport networks. We describe this evaluation methodology and present experimental results of key aspects of the PATN.

Closed-form expression for the collision probability in the IEEE Ethernet Passive Optical Network registration scheme

Abstract: We derive a closed-form expression for the message collision probability in the IEEE 802.3ah Ethernet Passive Optical Network (EPON) registration scheme. The expression obtained, although based on an approximation, shows a good match with simulation results. We use the results of our analysis to compute the size of the most efficient contention window and the most efficient number of nodes serviced by a given window size.