Overlay networks are becoming a popular vehicle for deploying advanced network services in the Internet. Overlay networks are implemented by deploying service nodes at suitably chosen sites in the network. Service nodes communicate with users through the commodity Internet, while among themselves, they may use either the commodity Internet or dedicated channels. The number of distinct service nodes has a big influence on the operational cost of an overlay network; meanwhile, the distance between service nodes and end users has a big influence on the quality of the service. In ...Read More

Overlay networks are becoming a popular vehicle for deploying advanced services in the Internet. One such service is multicast. Unlike conventional IP multicast, which requires universal deployment of network layer mechanisms, the overlay multicast model leverages the existing unicast mechanism and offers many service flexibilities to applications. Implementing multicast without requiring network support eliminates many deployment complexities that IP multicast has faced. However, it also raises new issues in efficient network design. In an earlier paper, we studied multicast routing algorithms designed to optimize resource usage in overlay networks, when ...Read More

Multicast is an efficient transmission scheme for supporting group communication in networks. Contrasted with unicast, where multiple point-to-point connections must be used to support communications among a group of users, multicast is more efficient because each data packet is replicated in the network – at the branching points leading to distinguished destinations, thus reducing the transmission load on the data sources and traffic load on the network links. To implement multicast, networks need to incorporate new routing and forwarding mechanisms in addition to the existing are not adequately supported in ...Read More

This thesis describes an unsupervised system to learn natural language morphology, specifically suffix identification from unannotated text. The system is language independent, so that is can learn the morphology of any human language. For English this means identifying “-s”, “-ing”, “-ed”, “-tion” and many other suffixes, in addition to learning which stems they attach to. The system uses no prior knowledge, such as part of speech tags, and learns the morphology by simply reading in a body of unannotated text. The system consists of a generative probabilistic model which is ...Read More

In recent years, networked sensors are finding use in a variety of different applications ranging from temperature monitoring to battlefield strategy planning. Advances in fabrication techniques have led to the development of sensor-actuator devices called MEMS. It has now become possible to move software closer to where the “action” is, i.e. the sensors themselves. These sensor devices typically have a micro-controller, instruction and data memory, a radio module for wireless communication and an operating system. These devices are severely resource constrained in terms of memory, processing power and energy, since ...Read More

Program visualization has the potential to be an important tool for people who seek to observe and understand the behavior of a running computation. This thesis focuses on alleviating barriers to the realization of this potential that pertain to the design of a visualization system and to insufficient knowledge about how people take advantage of program visualizations. Our major contribution is the design of a visualization approach capable of improving user’s performance through the use of economy of information and tasks. We present evidence from our empirical studies that this ...Read More

This is the final report For Washington University's Terabit Burst Switching Project, supported by DARPA and Rome Air Force Laboratory. The primary objective of the project has been to demonstrate the feasibility of Burst Switching, a new data communication service, which seeks to more effectively exploit the large bandwidths becoming available in WDM transmission systems. Burst switching systems dynamically assign data bursts to channels in optical datalinks, using routing information carried in parallel control channels.

The flexibility to adapt to new services and protocols without changes in the underlying hardware is and will increasingly be a key requirement for advanced networks. Introducing a processing component into the data path of routers and implementing packet processing in software provides this ability. In such a programmable router, a powerful processing infrastructure is necessary to achieve to level of performance that is comparable to custom silicon-based routers and to demonstrate the feasibility of this approach. This work aims at the general design of such programmable routers and, specifically, ...Read More

In this thesis we have studied the potential of Motion JPEG 2000 for video processing and compared its performance with current widely used video coding standards, Motion JPEG, MPEG-2, and MPEG-4. Four key aspects are compared among them, which are compression efficiency, error resilience, perceptual video quality, and computation complexity. Our experiments show that Motion JPEG 2000 provides high compression performance, strong error resilience, good perceptual video quality, and acceptable computation complexity for real-time video processing. Together with a rich set of features inherited from JPEG 2000, such as resolution ...Read More

Tracking or localizing a moving target is a difficult task in a distributed sensor network, due to the lack of knowledge of the target's motion and signal noises. Most existing approaches to the problem use only sensory information and may require accurate target's motion models. In this paper, we present a Markovian approach that combines dynamically estimated target's motion models with received sensory information. Without a given motion model, this approach localizes a target in two steps, a location prediction step using dynamically generated motion models and a location correction ...Read More

We study a resource allocation problem, which is a central piece of a real-world crew scheduling problem. We first formulate the problem as a hybrid soft constraint satisfaction and optimization problem and show that its worst-case complexity is NP-complete. We then propose and study a set of decision and optimization modeling schemes for the problem. We consider the expressiveness of these modeling schemes for the problem. We consider the expressiveness of these modeling methods. Specifically, we experimentally investigate how these modeling schemes interplay with the best existing systematic search and ...Read More

The ongoing increases of line speed in the Internet backbone combined with the need for increased functionality of network devices presents a major challenge. These demands call for the use of reprogrammable hardware to provide the required flexible, high-speed functionality, at all network layers. The Field Programmable Port Extender (FPX) provides such an environment for development of networking components in reprogrammable hardware. We present a framework to streamline and simplify networking applications that process ATM cells, AAL5 frames, Internet Protocol (IP) packets and UDP datagrams directly in hardware.

In response to the increasing network speeds, many operations in IP routers and similar devices are being made more efficient. With the advances in other areas of packet processing, the verification and regeneration of cyclic redundancy check (CRC) codes of the data link layer is likely to become a bottleneck in the near future. In this paper, we present a mechanism to defer CRC verification without compromising reliability. This opens the possibility of incremental updates of the CRC. We introduce a new high-speed technique and present efficient implementation, speeding up ...Read More

The ongoing exponential increase of line speed in the Internet and combined with the uncountable requests for increased functionality of network devices presents a major challenge. These demands call for the use of reprogrammable hardware to provide the required flexible high-speed functionaltiy. The Field Programmable Port Extender (FPX) provides such an environment for development of networking components in reprogrammable hardware. We present the high-speed IP routing components in reprogrammable hardware. We present the high-speed IP routing module "OBIWAN" (Optimal Binary search IP lookup for Wide Area Networks) built on top ...Read More

The provisioning of dynamic forms of services is becoming the main stream of today's network. In this paper, we focus on services assisted by network servers and different forms of associated sessions. We identify two types of services: transparent, where the session is unaware of the server location, and configurable, where the sessions need to be configured to use their closest server. For both types we formalize the problem of optimally placing network servers and introduce approximated solutions. We present simulation result of approximations and heuristics. We also solve the ...Read More

Scheduling problems arise each time there is some form of resource contention. The problem addressed by scheduling disciplines is that of ordering the access to contended resources. The ordering is typically based on either (1) properties that are exposed by the entities that compete for the resource (like a deadline), or by (2) external properties (like the arrival order), or (3) a combination of both. In literature there exist many different scheduling algorithms, each of which has certain properties and an associated application domain. All these scheduling disciplines are based ...Read More

This paper describes the architecture of the Smart Port Card (SPC) designed for use with the Washington University Gigabit Switch. The SPC uses an embedded Intel Pentium processor running open-source NetBSD to support network management and active networking applications. The SPC physically connects between a switch port and a normal link adapter, allowing cell streams to be processed as they enter or leave the switch. In addition to the hardware architecture, this paper describes current and future applications for the SPC.

Random Access Memory (RAM) is a common resources needed by networking hardware modules. Synchronous Dynamic RAM (SDRAM) provides a cost effective solution for such data storage. As the packet processing speeds in the hardware increase memory throughput can be a bottleneck to achieve overall high performance. Typically there are multiple hardware modules which perform different operations on the packet payload and hence all try to access the common packet buffer simultaneously. This gives rise to a need for a memory controller which arbitrates between the memory requests made by different ...Read More

One goal of any genome project is the elucidation of hte primary sequence of DNA contained within a given species. While the availability ot the primary sequence itself is valuable, it does not reach its full potential until i has been annotated. Generally defined, annotation is descriptive information or commentary added to text, in this case genomic sequence. Without a mechanism for collecting, recording, and disseminating community-based annotation, a valuable source of information is severely diminshed. In this report I outline the design and implementation of a Distributed Annotation System ...Read More

While hardware plugins are well suited for processing data with high throughput, software plugins are well suited for implementing complex control functions. A plugin module has been implemented for the FPX that executes software on an embedded soft-core processor. By including this module in an FPX design, it is possible to implement active networking functions on the FPX using both hardware and software. The KCPSM, an 8-bit microcontroller developed by Xilinx Corp., has been embedded into an FPX module. The module includes circuits to be reprogrammed over the network and ...Read More

The client-server model continues to dominate distributed computing with increasingly more flexible variants being deployed. Many are centered on the notion of discovering services at run time and on allowing any system component to act as a service provider. The result is a growing reliance on the service registration and discovery mechanisms. This paper addresses the issue of facilitating such service provision capabilities in the presence of (logical and physical) mobility exhibited by applications executing over ad hoc networks. The solution being discussed entailes a new kind of service model, ...Read More

Field Programmable Gate Arrays (FPGAs) can be partially reconfigured to implement Dynamically loadable Hardware Plugin (DHP) modules. A tool called PARBIT has been developed that transforms FPGA configuration bitfiles to enable DHP modules. With this tool it is possible to define a partial reconfigurable area inside the FPGA and download it into a specified region of the FPGA device. One or more DHPs, with different sizes can be implemented using PARBIT.

The design of ad hoc mobile applications often requires the availability of a consistent view of the application state among the participating hosts. Such views are important because they simplify both the programming and verification tasks. Essential to constructing a consitent view is the ability to know what hosts are within proximity of each other, i.e., form a group in support of the particular application. In this paper we propose a protocol that allows hosts within communication range to maintain a consistent view of the group membership despite movement and ...Read More

Packets in the Internet can experience large queueing delays during busy periods. Backbone routers are generally engineered to have large buffers, in which packets may wait as long as half a second (assuming FIFO service, longer otherwise). During congestion periods, these bufferfs may stay close to full, subjecting packets to long delays, even when the intrinsic latency of the path is relatively small. This paper studies the performance improvements that can be obtained by using more sophisticated packet schedulers, than are typical of Internet routers. The results show that the ...Read More

This paper describes the design, implementation, and performance of an open, high-performance, dynamically reconfigurable Multi-Service Router (MSR) being developed at Washington University in St. Louis. This router provides an experimentation platform for research on protocols, router software, and hardware design, network management, quality of service and advanced applications. The MSR has been designed to be flexible, without sacrificing performance. It support gigabit links and uses a scalable architecture suitable for supporting hundreds or even thousands of links. The MSR's flexibility makes it an ideal platform for experimental research on dynamically ...Read More

Soft constraint minimization problems (SCMPs) contain hard constraints that cannot be violated and soft constraints that may be violated but carry penalties if not satisfied. In this paper, we first extend local search, WalkSAT in particular, to SCMPs and study the existing SAT encoding schemes for SCMPs. We propose a general encoding method called k-encoding. We then investigate the effects of local search neiborhood structures introduced by encoding schemes and analyze the anytime performance of extended WalkSAT using different encoding methods. Our experimental results on various graph coloring problems show ...Read More

In this proposal, we consider the problem of resource reservation for Integrated Services (IntServ) and Differentiated Services (DiffServ) networks. Current approaches for resource reservation in INtegrated Service Networks adopt an all-or-nothing approach, where partially acquired resources must be released if resources are not available at all routers on the chosen path. Furthermore, under high load, end-systems must retry requests repeatedly leading to inefficient allocation and increased traffic. We propose a new approach called Deferred REServation (DERS) that substantially improves performance (reduces the overall cell rejection probability and increases link utilization) ...Read More

This paper studies the performance of deferred resource reservation in data networks. Conventional resource reservation protocols, such as PNNI and RSVP adopt an all-or-nothing approach, where partially acquired resources must be released if resources are not available at all links on the chosen path. During periods of high network load, this leads users to retry requests repeatedly, adding control traffic at exactly the time when the network's capacity to process that control traffic is exhausted. Deferred REServation (DRES) can significantly improve performance by reducing the overall call rejection probability, allowing ...Read More

To provide flexibility in deploying new protocols and services, general-purpose processing engines are being placed in the datapath of routers. Such network processors are typically simple RISC multiprocessors that perform forwarding and custom application processing of packets. The inherent unpredictability of execution time of arbitrary instruction code poses a significant challenge in providing QoS guarantees for data flows that compete for such processing resources in the network. However, we show that network processing workloads are highly regular and predictable. Using estimates of execution times of various applications on packets of ...Read More

Multi-dimensional scaling is an analysis tool which transforms pairwise distances between points to an embedding of points in space which are consistent with those distances. Two recent techniques in statistical patter recognition, locally linear embedding (LLE) and Isomap, give a mechanism for finding the structure underlying point sets for which comparisons or distances are only meaningful between nearby points. We give a direct method to extend the embedding algorithm to new topologies, finding the optimal embedding of points whose geodesic distance on a surface mathes the given pairwise distance measurements. ...Read More

The recent popularity of catadioptic and multi-camera imaging systems indicates a need to create formal models for general, non-perspective camera geometries. Development of algorithmic tools for interpreting images from a generalized camera model will lead to a better understanding of how to design camera systems for particular tasks. Here we define the corollary to epi-polar constraints for standard cameras - the relationship between two images of a scene taken by generalized cameras from viewpoints related by discrete or differential motions.

Video application over the Internet are getting increasingly popular because of the explosive growth of the Internet. However, video packets loss due to network congestions can degrade the video quality substantially. In this paper, we propose a transmission scheme for Motion-JPEG2000. It can be implemented in an active network environment efficiently. Our simulation shows that our scheme gracefully adapts to network congestion and improves the quality of video transmission in congested IP networks.

Optical Burst Switching (OBS) is an experimental network technology that enables the construction of very high capacity routers, using optical data paths and electronic control. In this paper, we study two designs for wavelength converting switches that are suitable for use in optical burst switching systems and evaluate their performance. Both designs use tunable lasers to implement wavelength conversion. One is strictly nonblocking design, that also requires optical crossbars. The second substitutes Wavelength Grating Routers (WGR) for the optical crossbars, reducing cost, but introducing some potential for blocking. We show ...Read More

Termination detection, a classical problem in distributed computing, is revisited in the new setting provided by the emerging mobile computing technology. A simple solution tailored for use in ad hoc networks is employed as a vehicle for demonstrating the applicability of formal requirements and design strategies to the new field of mobile computing. The approach is based on well understood techniquest in specification refinement, but the methodology is tailored to mobile applications and helps designers address novel concerns such as the mobility of hosts, transient interactions, and specific coordination constructs. ...Read More

Context-Aware computing is characterized by the ability of a software system to continuously adapt its behavior to a changing environment over which it has little or no control. Previous work along these lines presumed a rather narrow definition of context, one that was centered on resources immediately available to the component in question, e.g., communication bandwidth, physical location, etc. This paper explores context-aware computing in the setting of ad hoc networks consisting of numerous mobile hosts that interact with each other opportunistically via transient wireless interconnections. We extend the context ...Read More

As computing devices become smaller and wireless networking technologies improve, the popularity of mobile computing continues to rise. In today's business world, many consider devices such as cell phones, PDAs, and laptops as essential tools. As these and other devices become increasingly independent of the wired infrastructure, new kinds of applications that assume an ad hoc network infrastructure will need to be deployed. Such a setting poses new challenges for the software developer, e.g., the lack of an established network topology, bandwidth limitations, and frequent disconnections. In this paper, we ...Read More

We propose a new network and system architecture for multicast in the Internet. Our main objectives are to find a cost-effective way to scale to a large number of multicast groups whose members are geographically dispersed, and to enable small and less capable devices to participate in group communications. In order to preserve the efficiency of data distribution gained by multicast, while avoiding the control complexity previously exhibited by IP multicast, we propose the use of an overlay network for multicast services. We construct "virtual" multicast trees, which consist of ...Read More

Multicast servises can be provided either as a basic network service or as an application-layer service. Higher level multicast implementations often provide more sophisticated features, and since they don't require network supoprt for multicast, they can provide multicast services, where no network layer support is available. Overlay multicast networks offer an intermediate option, potentially combining the flexibility and advanced features of application layer multicast with the greater efficiency of network layer multicast. Overlay multicast networks play an important role in the Internet. Indeed, since Internet Service Providers have been slow ...Read More

The Field-programmable Port eXtender (FPX) provides dynamic, fast, and flexible mechanisms to process data streams at the ports of the Washington University Gigabit Switch (WUGS-20). By performing all computations in FPGA hardware, cells and packets can be processed at the full line speed of the transmission interface, currently 2.4 Gbits/sec. In order to design and implement portable hardware modules for the Reprogrammable Application Devide (RAD) on the FPX board, all modules should conform to a standard interface. This standard interface specifies how modules receive and transmit ATM cells of data ...Read More

The Field-programmable Port eXtender (FPX) provides dynamic, fast, and flexible mechanisms to process data streams at the ports of the Washington University Gigabit Switch (WUGS-20). In order to facilitate the design and implementation of portable hardware modules for the Reprogrammable Application Device (RAD) on the FPX board, infrastructure components have been developed. These components abstract application module designers from device-specific timing specifications of off-chip memory devices, as well as processing system-level control cells. This document describes the design and internal functionality of the infrastructure components and is intended as a ...Read More

Continuing growth in optical link speeds places increasing demands on the performance of Internet routers, while deployment of embedded and distributed network services imposes new demands for flexibility and programmability. IP adress lookup has become a significant performance bottleneck for the highest performance routers. New commercial products utilize dedicated Content Addressable Memory (CAM) devides to achieve high lookup speeds. This paper describes an efficient, scalable lookup engine design, able to achieve high-performance with the use of a small portion of a reconfigurable logic device and a commodity Random Access Memory ...Read More

Users and developers of visualization tools must deal with the problem of specifying what information to show and how to represent it. Typically, the user's focus of interest will change over time, and the specifications must change with the user's interests. Techniques for the simple, direct, and intuitive creation and refinement of these specifications can be useful. In this paper we show how legends, a natural element of graphical displays, may be used as a direct and unobstrusive interaction device through which users may interactively specify new visualizations and animations.

This report summarizes progress on Washington University's Terabit Burst Switching Project, supported by DARPA and Rome Air Force Laboratory. This project seeks to demonstrate the feasibility of Burst Switching, a new data communication service which can more effectively exploit the large bandwidths becoming available in WDM transmission systems, than conventional communication technologies like ATM and IP-based packet switching. Burst switching systems dynamically assign data bursts to channels in optical data links, using routing information carried in parallel control channels. The project will lead to the construction of a demonstration switch ...Read More

This report summarizes progress on Washington University’s Terabit Burst Switching Project, supported by DARPA and Rome Air Force Laboratory. This project seeks to demonstrate the feasibility of Burst Switching, a new data communication service which can more effectively exploit the large bandwidths becoming available in WDM transmission systems, than conventional communication technologies like ATM and IP-based packet switching. Burst switching systems dynamically assign data bursts to channels in optical data link, using routing information carried in parallel control channels. The project will lead to the construction of a demonstration switch ...Read More

When using an on-demand media streaming system on top of a network with Multicast support, it is sometimes more efficient to use broadcast to distribute popular content. There has been a lot of research in broadcasting on-demand content to multiple, asynchronous receivers. In this paper, we propose a family of novel, practical techniques for broadcasting on-demand media, which achieve lowest known server/network bandwidth usage and I/O efficient client buffer management, while retaining the simplicity of a frame-based single channel scheme.

The growing deployment of wireless local area networks (WLANs) in corporate environments, the increasing number of wireless Internet service providers and demand for quality of service support create a need for controlled sharing of wireless bandwidth. In this technical report a method for using long-term channel-state information in order to control the sharing of wireless bandwidth is studied. The algorithm enables an opera-tor of a WLAN to equalize the revenue/cost for each customer in the network and to control the link-sharing based on a combination of user-centric and operator-centric fact-tors. ...Read More

Active Networking is the basis for a range of new and innovative applications that make use of computational resources inside network routers. One such application is Aggregated Hierarchical Multicast, which aims at implementing efficient many-to-many communication. In certain scenarios it is possible to transmit less accurate, aggregated data and thus achieve better scalability. Using Active Networks, this aggregation computation can be done transparently by network routers without end system support. We present how aggregated data streams can be structured in a hierarchical fashion to allow easy access of data at ...Read More

The new ISO/ITU-T standard for still image coding, JPEG2000, has been shown to provide superior coding efficiency to the previous standard, JPEG. Because of the superb performance of JPEG2000, it is reasonable to argue that Motion-JPEG2000, the corresponding moving picture coding standard of JPEG2000, has equally outstanding performance. However, there has not been a sufficient performance evaluation of Motion-JPEG2000. To this end, we have studied the potential of Motion-JPEG2000 for video processing. Our experiments show that Motion-JPEG2000 provides high compression performance, strong error resilience, and good perceptual image quality. Together ...Read More

While advances in computer and communications technology have made the network ubiquitous, they ahve also rendered networked systems vulnerable to malicious attacks orchestrated from a distance. These attacks, usually called cracker attacks or intrusions, start with crackers infiltrating a network through a vulnerable host and then going on to launch further attacks. Crackers depend on increasingly sophisticated techniques like using distributed attack sources. On the other hand, software that guards against them remains rooted in traditional centralized techniques, presenting an easily-targetable single point of failure. Scalable, distributed network intrusion prevention ...Read More

Many real-world problems involve constraints that cannot be all satisfied. The goal toward an overconstrained problem is to find solutions minimizing the total number of constraints violated. We call such a problem constraint minimization problem (CMP). We study the behavior of the phase transitions and backbones of CMP. We first investigate the relationship between the phase transitions of Boolean satisfiability, or precisely 3-SAT (a well-studied NP-complete decision problem), and the phase transitions of MAX 3-SAT (an NP-hard optimization problem). To bridge the gap between the easy-hard-easy phase transitions of 3-SAT, ...Read More