Michael S. Borella

Optical components for WDM lightwave networks

Recently, there has been growing interest in developing optical fiber networks to support the increasing bandwidth demands of multimedia applications, such as video conferencing and World Wide Web browsing. One technique for accessing the huge bandwidth available in an optical fiber is wavelength-division multiplexing (WDM). Under WDM, the optical fiber bandwidth is divided into a number of nonoverlapping wavelength bands, each of which may be accessed at peak electronic rates by an end user. By utilizing WDM in optical networks, we can achieve link capacities on the order of 50 THz. The success of WDM networks depends heavily on the available optical device technology. This paper is intended as a tutorial on some of the optical device issues in WDM networks. It discusses the basic principles of optical transmission in fiber and reviews the current state of the art in optical device technology. It introduces some of the basic components in WDM networks, discusses various implementations of these components, and provides insights into their capabilities and limitations. Then, this paper demonstrates how various optical components can be incorporated into WDM optical networks for both local and wide-area applications. Finally, the paper provides a brief review of experimental WDM networks that have been implemented.

Efficient scheduling of nonuniform packet traffic in a WDM/TDM local lightwave network with arbitrary transceiver tuning latencies

A passive-star-based, broadcast-and-select, local lightwave network which can support a limited number of wavelength-division multiplexed (WDM) channels, but serve a much larger number of nodes, is considered. Each node is equipped with one tunable transmitter and one fixed receiver, and each WDM channel is operated in a time-division multiplexed (TDM) fashion for carrying packet traffic. Bandwidth is allocated to the node pairs when traffic flow between them is nonuniform, while also accommodating transceiver tuning latency. Our approach exploits well-known results from scheduling theory to create efficient transmission schedules. Multiprocessor task scheduling heuristics that can be applied to load balancing in a multichannel network is also examined.

Internet packet loss: measurement and implications for end-to-end QoS

We analyze a month of Internet packet loss statistics for speech transmission using three different sets of transmitter/receiver host pairs. Our results indicate that packet loss is highly bursty, with the majority of individual losses occurring in a relatively small number of bursts. We find that loss exhibits dependence in most cases, but is not always well modeled as dependent. We introduce an analytical technique for measuring loss dependency. We also consider the asymmetry of round trip packet loss, and find that most loss on a round trip path occurs in either one direction or the other. We introduce a normalized metric for measuring loss asymmetry and apply it to our measurements. Finally we discuss the implications of our study for the next generation of real time voice services in the Internet.

A reservation-based multicasting protocol for WDM local lightwave networks

A wavelength-division multiplexing (WDM) based multicasting protocol for a single-hop broadcast-and-select local lightwave network is proposed. The approach employs a control-channel-based media-access protocol that schedules multicast packets while incorporating arbitrary transceiver tuning times and propagation delays. An arbitrary number of data channels (W) supply communication bandwidth to N nodes, where N/spl ges/W. An additional control channel is used for synchronization and scheduling. Each node is equipped with one fixed transmitter and one fixed receiver on the control channel, as well as one tunable (wavelength-agile) transmitter and one or more tunable receivers for data channel access. Results suggest the WDM single-hop multicasting experiences very good performance when multicast size is either small or very large, or when nodes are equipped with multiple receivers.

Internet delay effects: how users perceive quality, organization, and ease of use of information

In this paper we report the results of an investigation designed to determine the effects of Internet delays on users perceptions of ease of locating information, organization of information, quality of information, and navigation problems. The results demonstrated user sensitivity to delays. As expected, for text-and-graphics documents, shorter delays provoked more favorable responses. However, for text-only documents, the shorter the delay, the less favorably a document was viewed. The results indicated that users may prefer multi-media web sites but are unwilling to tolerate the substantial network delays often associated with delivering graphics, video, animation, and audio.

Self-similarity of Internet packet delay

In this paper we present results which suggest that Internet packet delay, when viewed as a time series, is self-similar in nature. A self-similar phenomenon displays the same or similar statistical properties across a wide range of time scales. Self-similarity has previously been observed in the magnitude of traffic transmitted on local-area and wide-area networks. Our research is the first to explore the self-similarity of traffic delay rather than traffic magnitude. Our results imply that the delay experienced by application layer protocols is extremely bursty and that traditional Poisson models cannot accurately predict the degree of burstiness. Furthermore, we present evidence that the degree of self-similarity for a round-trip path in the Internet may be correlated with the packet loss observed on that path.

Pricing Framework for a Differential Services Internet

A number of recent proposals and proposed standards have addressed adding differential services to the Internet. Although their details and tentative implementations differ, most are recommending what essentially amounts to multiple levels of best-effort service. In this paper, we survey recent differential services and pricing proposals and introduce a pricing framework for a differentiated-services network that focuses on simplicity, flexibility, and ease of implementation. In particular, our model can be used in the current heterogeneous Internet without major restructuring. We focus on flat-rate, per-time and usage-based pricing, where users can change their service level on an ad-hoc basis. We also show that a sender-pays model with back-charging is simple to implement and can be effective even in complicated transactions, such as multicasting. Our approach is different from many previous proposals and consciously integrates differential services and pricing with implementation as the immediate goal. Finally, we outline future areas of research including ISP support for pricing, the dynamics of service quality and pricing in a differentiated Internet, and the latest trends and directions of QoS deployment and standardization.

Limits of Multicasting in a Packet-Switched WDM Single-Hop Local Lightwave Network

Wavelength division multiplexing (WDM) is a technique used to divide the tremendous optical bandwidth of a fiber into many non-interfering channels. A WDM single-hop local lightwave network can be designed by using two-way fibers to connect every node to a passive star coupler, and equipping the nodes with tunable transmitters and/or receivers. The passive star broadcasts data transmitted on each wavelength channel to all nodes in the system. This type of network, known as broadcast and select, is naturally suited to multicasting; every node in the network has the opportunity to receive any particular packet. In this study, we show how multicasting in a broadcast and select environment can dramatically increase the throughput of the network. Using a Markov model and a simulation, we show that multicasting significantly improves the limiting performance of both small and large networks, especially when each node is equipped with multiple receivers.

Simple Scheduling Algorithms for Use with a Waveguide Grating Multiplexer Based Local Optical Network

As the need for greater bandwidth in local-area networks grows, wavelength-division multiplexing (WDM) is gathering attention as a viable successor to Gigabit Ethernet technologies. In this paper we introduce a new WDM optical LAN architecture based on the waveguide grating multiplexer (WGM) rather than the traditional passive star coupler (PSC). An N-port WGM allows N2 × N2 connectivity via only N physical wavelengths, due to its inherent space-division multiplexing property. Wavelength-routed networks based on the WGM promise to be significant components of future LAN and WAN technologies not only because of their efficient use of wavelengths, but also because they have been implemented as integrated devices. We propose simple, low-complexity TDM schedules for interconnecting MN nodes (M nodes per port) with a WGM in a local environment. Each node is equipped with a single tunable transmitter and a single tunable receiver (both of which can access all N wavelengths). Various transmitter and receiver tuning latencies are considered. We show that, for negligible tuning latencies, aggregate network throughput approaching min(MN, N2) can be achieved, and for tuning latencies on the order of a packet length or more, throughput on the order of N can be achieved. Since these performance metrics are vastly superior to that of an equivalent PSC-based system (whose maximum throughput is limited by the number of wavelengths, N), we propose that the WGM be considered as an alternative to the PSC for enabling WDM LANs and multiprocessor interconnects.

WANDS: Wide-Area Network Delay Simulator

Designing documents that will be viewed from remote locations via a network requires an understanding of traditional document and interaction design issues, plus an understanding of how network delays will impact document delivery. The paper describes a set of tools that allow designers to view documents stored locally while experiencing response time delays as if the documents were delivered from a different location on a network. By using measured network latencies to drive an instrumented World-Wide Web server, designers can view the documents they create from the perspective of someone sitting down the hall, across the country, or across an ocean.