Showing papers in "Journal of Optical Networking in 2004"

Optical-packet-switched interconnect for supercomputer applications[Invited]

Abstract: Feature Issue on Optical Interconnection Networks (OIN) We describe a low-latency, high-throughput scalable optical interconnect switch for high-performance computer systems that features a broadcast-and-select architecture based on wavelength- and space-division multiplexing Its electronic control architecture is optimized for low latency and high use Our demonstration system will support 64 nodes with a line rate of 40 Gbit/s per node and operate on fixed-length packets with a duration of 512 ns using burst-mode receivers We address the key system-level requirements and challenges for such applications

Light-trails: an optical solution for IP transport (Invited)

Abstract: Feature Issue on Next-Generation WDM Network Design and Routing (WDMN). We present a solution for IP-centric communication at the optical layer through a combination of a hardware platform and algorithmic implementation. The presented approach, termed light-trails, is shown to yield a reconfigurable networking platform in which optical connections of arbitrary duration can be established and torn down flexibly in negligible time, accommodating the dynamic traffic requirements of the IP world. The hardware platform and protocol are evaluated with tractable mathematical models validated through detailed simulation.

Ultrafast optoelectronic packet processing for asynchronous, optical-packet-switched networks [Invited]

Abstract: Feature Issue on Optical Interconnection Networks (OIN). We describe hybrid optical-electrical systems that perform header processing and buffering of ultrafast, asynchronous optical packets. Our systems are enabled by three key, novel devices: an all-optical serial-to-parallel converter, an optical clock-pulse generator, and a photonic parallel-to-serial-converter. These devices allow utilization of complementary metal-oxide semiconductor technology for compact, highly functional optical packet processing. A simplified node architecture for asynchronous, optical- packet-switched networks is made possible by these systems with all the necessary node functions integrated compactly. We also demonstrate an optical label swapper and a photonic random access memory for 40-Gbit/s, 16-bit, asynchronous optical packets.

Optical performance monitoring techniques based on pilot tones for WDM network applications

Abstract: Feature Issue on Optical Performance Monitoring (OPM). We report on the pilot-tone-based optical performance monitoring techniques for use in the wavelength-division multiplexing (WDM) network. We first described the operating principle of the pilot-tone-based monitoring technique and estimated its scalability by analyzing potential problems. We then reviewed the techniques proposed for monitoring various optical parameters of WDM signals--such as optical power, wavelength, optical path, and cross talk--using pilot tones. We also presented examples of various network elements using the pilot-tone-based monitoring technique. In addition, we reviewed the pilot-tone-based monitoring techniques used in adaptive compensators for chromatic dispersion and polarization-mode dispersion, and we discussed their limitations and possible solutions. For example, we demonstrated a simple technique for monitoring the chromatic dispersion and polarization-mode dispersion simultaneously, without their affecting each other, by scrambling the state of polarization of the optical signal.

Traffic grooming for survivable WDM networks: dedicated protection [Invited]

Abstract: We investigate the survivable traffic-grooming problem for optical mesh networks employing wavelength-division multiplexing (WDM) and dedicated protection. We consider the dynamic-provisioning environment in which a connection arrives at random, holds for a random amount of time, and then departs. A typical connection request may require bandwidth less than that of a wavelength, and it may also require protection from network failures, typically fiber cuts. On the basis of generic grooming-node architecture, we propose two approaches--protection-at-lightpath (PAL) level and protection-at-connection (PAC) level--for grooming a connection request. Here we investigate dedicated protection. In a companion paper [IEEE J. Sel. Areas Commun. 21, 1367 (2003)], we investigate shared protection, which leads to a substantially different treatment. For dedicated protection, we prove that the problem of provisioning a connection under PAC is NP-complete, propose effective heuristics for both schemes, and define comprehensive performance metrics to compare PAL with PAC with respect to wavelength or grooming-port efficiency. Our findings are as follows. Under today's typical connection-bandwidth distribution in which lower-bandwidth connections outnumber higher-bandwidth connections, PAC outperforms PAL (in terms of bandwidth-blocking ratio, lightpath use, and wavelength use) if the number of grooming ports is large; however, PAL outperforms PAC (in terms of bandwidth-blocking ratio and grooming-port use) when the number of grooming ports is moderate or small.

Recent progress in the scheduling algorithms in optical-burst-switched networks [Invited]

Abstract: Feature Issue on Next-Generation WDM Network Design and Routing (WDMN). Optical burst switching (OBS) is a promising paradigm for the future Internet infrastructure. Owing to the lack of an optical buffer, burst loss in OBS networks is a major concern. The problem of how to schedule as many bursts as possible to minimize the burst loss has been under intense research in the past several years. We classify the approaches to burst scheduling and review the recent research.

Improving fairness in optical-burst-switching networks

Abstract: Approaches for dealing with the beat-down unfairness problem have been mainly limited to electronic packet-switching networks. We investigate two methods for alleviating this problem in optical-burst-switching (OBS) networks. When OBS is used with just-in-time or just-enough-time scheduling, the control packet transmitted ahead of the data burst may fail to secure a free channel, causing the data burst to be dropped at some node along its path. The burst dropping probability increases as the number of hops in the lightpath of the burst increases, resulting in unfairness to bursts with long paths. Our first scheme to deal with this problem is based on a simple equation that adjusts the size of the search space for a free wavelength based on the number of hops traveled by the burst. Our second scheme uses proactive discarding to reduce the probability of dropping bursts with large hop count at the expense of an increase in the dropping probability of bursts with small hop count. The scheme adapts the random-early-discard (RED) strategy to the OBS environment and prioritizes the levels of discarding on the basis of the length of the lightpath. The proactive burst discarding is done in the network access station (NAS) of the source node that generated the burst. Thus discarded bursts do not waste any bandwidth resources in the core of the optical network. Both schemes have simple logic, do not use preemption, do not require complex modification to the lightpath setup scheme, and alleviate the beat-down unfairness without negatively affecting the overall throughput of the system. Detailed performance evaluation of the two schemes is presented and analyzed.

Wavelength converter placement in least-load-routing-based optical networks using genetic algorithms

Abstract: Feature Issue on Next-Generation WDM Network Design and Routing (WDMN). We study the problems of routing and wavelength converter placement in optical networks with sparse wavelength conversion. We propose a new dynamic routing algorithm with two new path cost functions based on the concept of least-load routing (LLR) with sparse converter placement, and we discuss the application of genetic algorithms (GAs) to determine the optimal location of wavelength converters so that the call-blocking probability is minimized. Simulation results show that the proposed dynamic routing algorithms perform significantly better than shortest-path (SP) routing and fixed-alternative routing (FAR) in terms of the call-blocking probability. The GA model is able to obtain a nearly optimal solution of the wavelength converter placement problem within a reasonable time, and its performance is better than that of two other popular heuristic placement algorithms.

Semiconductor optical amplifiers: performance and applications in optical packet switching (Invited)

Abstract: Semiconductor optical amplifiers (SOAs) are a versatile core technology and the basis for the implementation of a number of key functionalities central to the evolution of highly wavelength-agile all-optical networks. We present an overview of the state of the art of SOAs and summarize a range of applications such as power boosters, preamplifiers, optical linear (gain-clamped) amplifiers, optical gates, and modules based on the hybrid integration of SOAs to yield high-level functionalities such as all-optical wavelength converters/regenerators and small space switching matrices. Their use in a number of proposed optical packet switching situations is also highlighted.

Design alternatives for optical-packet-interconnection network architectures[Invited]

Abstract: Feature Issue on Optical Interconnection Networks (OIN). In the design of an optical-packet-switching network, several policy decisions must be made and network parameters must be dimensioned. These decisions determine the network operation and performance. We present an overview of the design alternatives for optical-packet-interconnection network architectures. The issues that are discussed include contention resolution schemes, header processing, and quality-of-service support. The design of metropolitan-area packet-switched networks is also presented in a separate section.

Wavelength conversion algorithm in an intelligent optical network from a multilayered approach

Abstract: We aim to implement a new deterministic wavelength conversion algorithm in local area networks (LANs). The network has been studied from a layered approach, and we describe how it can be interpreted as both a trunk network and a LAN. The limitation on the total number of wavelength conversions that a packet can undergo on its path has been addressed. Our algorithm employs online routing methods to attribute intelligence to the network and thus provides an improvement in the call connection probability.

Interdomain dynamic wavelength routing in the next-generation translucent optical Internet

Abstract: Feature Issue on Next-Generation WDM Network Design and Routing (WDMN). Future home and enterprise Internet users need high-bandwidth end-to-end connections that have to be provisioned automatically and dynamically. We need to make optical-layer routing as scalable and flexible as IP routing in today's Internet to support high-end applications. We study the problem of interdomain dynamic wavelength routing based on a multidomain translucent optical network model, which allows end-to-end lightpaths to be set up not only across multiple routing domains but also through two network layers: the optical layer and the regeneration layer. In this model, a lightpath can traverse the domain boundary either through optical bypass (OOO) or through optical-electrical-optical regeneration (OEO). We propose an interdomain dynamic wavelength-routing scheme with modest computational complexity to address the problem from an algorithmic perspective. Our experimental results show that the proposed interdomain dynamic wavelength-routing scheme can (a) flexibly incorporate various cost metrics and local (or intradomain) routing schemes into a uniform interdomain routing framework, (b) effectively set up end-to-end lightpaths in a multidomain translucent network, and (c) efficiently use both optical- and regeneration-layer resources.

High-speed optical time-division-multiplexed/WDM networks and their network elements based on regenerative all-optical ultrafast wavelength converters

Abstract: We describe an optical time-division-multiplexed (OTDM)/WDM network architecture that integrates high-speed optical time-division multiplexing at speeds of 40 Gbit/s and higher with lower-bit-rate WDM channels. An ultrafast wavelength converter is used as a regenerative multifunction building block for OTDM multiplexing, WDM-to-OTDM and OTDM-to-WDM transmultiplexing, OTDM/WDM multicasting, OTDM all-optical label read-write, and all-optical time-channel add-drop multiplexing. Subsystem design and underlying component technologies are described in detail. New results and performance measurements are shown at 40 and at 80 Gbit/s.

Broadcast polling--an uplink access scheme for the Ethernet passive optical network

Abstract: An uplink access scheme for the Ethernet passive optical network (EPON), called broadcast polling (BP), has been proposed. In the BP scheme, the optical line terminal (OLT) knows all optical network units' (ONUs) bandwidth requirements before allocating bandwidth to them, and only one gate message is used to inform all ONUs of their bandwidth grant information in every cycle. Compared with interleaved polling with adaptive cycle time (IPACT)--an important access scheme--BP can adopt more- sophisticated bandwidth-allocation algorithms so that it can better support the service-level agreement (SLA). Based on packet-delay analysis for the BP scheme, a simple bandwidth-allocation algorithm aiming to provide three different service levels to customers has been designed. The simulation results show that the algorithm works well. In fact, more-sophisticated algorithms can be designed on the basis of the BP scheme to support more-sophisticated SLAs.

Investigation of the scalability of dynamic wavelength-routed optical networks[Invited]

Abstract: Feature Issue on Optical Interconnection Networks (OIN). We describe results of the scalability analysis for dynamic wavelength-routed optical networks with end-to-end lightpath assignment and central network control with electronic scheduling and processing of lightpath requests. We investigate the effect of the algorithm complexity in both the scheduling and the dynamic routing and wavelength assignment (DRWA) of lightpath requests. Scheduling theory and static performance-prediction techniques were applied to define the bounds on the electronic processing time of requests, and hence the maximum number of nodes supported by a centralized dynamic optical network for given blocking probability, latency, and network diameter. Scalability analysis results show that medium-sized centralized networks (~50 nodes) can be supported when these networks are reconfigured on a burst-by-burst basis. In addition, we found that real topologies showed a complex trade-off between the request processing time, blocking probability, and resource requirements. These findings can be used to determine the optimum combination of scheduling/DRWA algorithm, showing that the fastest DRWA algorithm does not necessarily lead to the minimum blocking probability and maximum scalability but that a careful consideration of both blocking and processing speed is required. The results are applicable both to dynamic network architectures with centralized request processing such as wavelength-routed optical networks and to the design of advanced optical switching matrices and routers.

All-optical signal processing for optical packet switching

Abstract: Feature Issue on Optical Interconnection Networks (OIN). We present three optical signal processing functional blocks that enable 1×N optical packet switching. An ultrafast asynchronous multioutput all-optical header processor is demonstrated with a terahertz optical asymmetric demultiplexer in combination with a header preprocessor. It is shown that self-induced polarization rotation can be used for both the header processor and the header preprocessor. The second functional block is optical buffering. This is shown with both a laser neural network and a recirculating buffer. Related to this is a three-state all-optical memory based on coupled lasers, which increases the number of possible output states of an optical packet switch.

Performance of an OSNR monitor based on the polarization-nulling technique

Abstract: Feature Issue on Optical Performance Monitoring (OPM). We report on the performance of an optical signal-to-noise ratio (OSNR) monitor with the polarization-nulling technique. The results show that this monitor was robust to polarization-mode dispersion and in-band cross talk. For a demonstration, we monitored the OSNR of the optical signal transmitted through a 2×60 km long aerial fiber link for one week and confirmed that the OSNR could be measured with accuracy better than ±0.3 dB (when the OSNR is lower than 26 dB).

Does header length affect performance in optical burst switched networks

Abstract: Feature Issue on Next-Generation WDM Network Design and Routing (WDMN). We investigate the effect of nonnegligible header length (HL) in optical burst switching on blocking probability. The HL is the total delay of a control packet at the controller. We first develop a model that explicitly presents the distribution of offset times as a function of the HL. Next we argue that the variance of this distribution (and not the mean) affects the blocking probability. In particular, the total blocking probability of a burst is dominated by the blocking on its last link, where its offset is shortest. We derive a lower bound for a HL threshold value below which blocking is not sensitive to the reservation algorithm. This threshold depends on network connectivity, number of channels per fiber, and burst length. The blocking probabilities of both the just enough time and the horizon reservation algorithms were empirically found not to be very sensitive to the distribution of burst sizes.

Emulation of realistic network traffic patterns on an eight-node data vortex interconnection network subsystem

Abstract: Feature Issue on Optical Interconnection Networks (OIN). We demonstrate the feasibility of the data vortex interconnection network architecture for use in supercomputing by emulating realistic network traffic on an eight-node subnetwork. The evaluation workload uses memory accesses from the Barnes-Hut application in the SLPASH-2 parallel computing benchmark suite, which was extracted by using the M5 multiprocessor system simulator. We confirm that traffic is routed correctly and efficiently.

Packet loss rate and jitter differentiating Quality-of-Service schemes for asynchronous optical packet switches

Abstract: Feature Issue on Optical Interconnection Networks (OIN). We propose access-restriction-based quality-of-service differentiation schemes, suitable for an asynchronous optical packet switch with a contention resolution pool that contains both tunable wavelength converters and fiber delay lines. The schemes aim at obtaining a high degree of packet-loss-rate isolation, for a low increase in overall packet loss rate, at the same time respecting the jitter tolerance of each class of service. Numerical simulations quantify how the performance depends significantly on the jitter tolerance of the traffic in general, and of the highest-priority class of service in particular.

Resource-efficient path-protection schemes and online selection of routes in reliable WDM networks

Abstract: Feature Issue on Next-Generation WDM Network Design and Routing (WDMN). The optimal choice of routing and wavelength assignment (RWA) for the working and protection path-pair of the newly generated demand request is often a complex problem in reliable wavelength-division-multiplexed (WDM) networks subject to dynamic traffic. The challenge is twofold: how to provide the required reliability level without over-reserving network resources and how to find a good solution of the RWA problem under constrained computational time. Two important contributions are made. First, the shared path protection (SPP) switching scheme is generalized to guarantee the required (differentiated) level of reliability to all arriving demands, while, at the same time, ensuring that they contain the required amount of reserved network resources. This generalization is referred to as SPP-DiR. Second, an approach for choosing the working and protection path-pair routing for the arriving demand is proposed. The approach is based on a matrix of preselected path-pairs: the disjoint path-pair matrix (DPM). Results show that, when the SPP-DiR scheme is applied, a small reduction in demand reliability corresponds to a significant reduction of the required network resources, when compared with the conventional SPP. In turn, the demand blocking probability may be reduced more than one order of magnitude. It is also shown that the DPM approach is suitable for obtaining satisfactory RWA solutions in both SPP-DiR and conventional SPP networks. The use of the DPM is most suited when the time for solving the RWA problem is constrained, e.g., when demand requests must be served swiftly.

Asynchronous sampled amplitude histogram models for optical performance monitoring in high-speed networks

Abstract: Feature Issue on Optical Performance Monitoring (OPM). A methodology that allows us to characterize the main performance parameters of high-rate (up to 40 Gbit/s) systems based on slow, bit-rate-independent sampling is presented taking into consideration all the information present in the histogram. The theory and validating simulation and experimental results are presented, showing that this technique is suitable for future transparent optical networks.

Active light-path restoration in WDM networks [Invited]

Abstract: Feature Issue on Next-Generation WDM Network Design and Routing (WDMN). The increasing demand for bandwidth that has resulted from the emergence of various bandwidth-thirsty applications has increased the need for employing WDM as a dominant technology for the next-generation optical internet. As these networks carry gigantic amounts of information, survivability has become a primary and basic need for them. Although the available proactive protection schemes provide a 100% restoration guarantee, they require huge redundancy and usually result in a very high blocking probability. We propose a novel active restoration scheme that restores a failed primary lightpath based on multiple predefined but nonreserved backup paths along the primary path. Extensive simulation results based on three typical test networks indicate that with a small increase in restoration time, our new restoration scheme can reduce the blocking probability and the capacity requirement significantly while guaranteeing a very high restoration probability.

Implementation of medium access control for interconnecting slotted rings to form a WDM metropolitan area network

Abstract: Feature Issue on Optical Interconnection Networks (OIN) A system of slotted interconnected rings employing a combination of wavelength-division multiple access (WDMA) with time-division multiple access (TDMA) can serve a metropolitan area without electro-optical conversion and buffering of payload except at system entry points The multiple rings overcome the power budget limitations of the single ring extending the reach of the system to even the largest metropolitan areas, the WDM dimension provides flexibility and ease of evolution, and the TDMA dimension offers the efficiency of multiplexing gain particularly under bursty traffic The system control information is transferred on a dedicated wavelength and is processed in the electrical domain at the ring nodes and the hub, which interconnects the rings The algorithms control the access to each ring and the scheduling of slots among the rings, based on explicit reservations, to adapt efficiently to the fluctuating offered load We present the design and hardware implementation of the access control algorithms for such a system built in the framework of the Information Society Technologies (IST) project DAVID (data and voice over DWDM)

Optoelectronic router with a reconfigurable shuffle network based on micro-optoelectromechanical systems

Abstract: An optoelectronic router with a shuffle exchange network is presented and enhanced by the addition of micro-optoelectromechanical systems (MOEMS) in the network to add the ability to reconfigure the shuffle network. The MOEMS described here are fully connected any-to-any crossbar switches. The added reconfigurability provides the opportunity to adapt the system to different common application characteristics. Two representative application models are described: The first has symmetric properties, and the second has asymmetric properties. The router system is simulated with the specified applications and an analysis of the results is carried out. By use of MOEMS in the optical network, and thus reconfigurability, greater than 50% increased throughput performance and decreased average packet delay are obtained for the given application. Network congestion is avoided throughout the system if reconfigurability is used.

Fair uni- and multicasting in a ring metro WDM network

Abstract: Packet-switched wavelength-division multiplexing (WDM) ring networks have been extensively studied as solutions to the increasing amount of traffic in metropolitan area networks, which is widely expected to be a mix of unicast and multicast traffic. We study the fairness between unicasting and multicasting in slotted packet-switched WDM ring networks that employ a tunable transmitter and fixed tuned receiver at each node and a posteriori buffer selection. We find that single-step longest-queue (LQ) buffer selection generally results in unfairness between unicasting and multicasting or a fixed relative priority for multicast versus unicast traffic. We propose and evaluate dual-step buffer selection policies that achieve fairness and allow for a range of relative priorities of multicast versus unicast traffic.

Alternative structures for two-dimensional MEMS optical switches (Invited)

Abstract: Feature Issue on Optical Interconnection Networks (OIN). Two-dimensional (2-D) microelectromechanical system (MEMS) optical switches have the merits of easy fabrication and high reliability. Since the optical signal loss is mainly proportional to the length of signaling paths in the switches, current 2-D MEMS optical switches that use a crossbar structure have a rather limited number of ports. For larger 2-D MEMS optical switches, we may use nonrectangular topology switching fabrics to shorten the internal signaling path or to recollimate the optical signal segment by segment inside the switches. We discuss these approaches from the aspect of implementation and routing control complexity.

Q-factor monitoring using FEC for fault-management applications

Abstract: Feature Issue on Optical Performance Monitoring (OPM). We examine the use of forward-error-correction coding for Q-factoring monitoring in fault-management applications. The sensitivity of this technique is measured for accumulated dispersion values up to -982 ps/nm, demonstrating the potential for use in embedded monitoring without dispersion compensation.

All-optical wavelength conversion for optically multiplexed 40-Gbit/s payload and 2.5-Gbit/s label in an optical-label-switched network

Abstract: Wavelength conversion for an optical packet with an optically multiplexed 40-Gbit/s payload and a 2.5-Gbit/s label has been demonstrated for the first time to our knowledge in an optical-label-switched network. The power penalties caused by wavelength conversion for the 40-Gbit/s payload and the 2.5-Gbit/s label are 1.1 and 2.3 dB, respectively. We have also demonstrated that the optical packet is successfully transmitted over two hops with a record transmission distance of 200 km.

Priority scheme for supporting quality of service in optical burst switching networks

Abstract: Feature Issue on Optical Interconnection Networks (OIN). We present what we believe to be a new scheme, called preemptive prioritized just enough time (PPJET), for quality-of-service (QoS) provisioning in bufferless optical burst switching (OBS) networks. PPJET provides better service for high-priority traffic by dropping reservations belonging to lower-priority traffic with a new channel-scheduling algorithm called preemptive latest-available unused channel with void filling (PLAUC-VF). An approximate method for calculating the dropping probability in PPJET is also discussed. Moreover, we perform discrete-event simulations to evaluate the performance of PPJET. Simulation results show that PPJET outperforms offset-based QoS schemes both in terms of dropping probability and end-to-end delay.