All Computer Science and Engineering Research

This report summarizes progress on the Terabit Burst Switching Project at Washington University for the period from December 15, 1997 through March 15, 1998. Efforts during this period have concentrated on working out details of the burst switch architecture, evaluating a variety of implementation alternatives and developing the physical design of the 160 Gb/s ATM switch to allow demonstration of the burst switch within a realistic network context.

This report summarizes progress on the Terabit Burst Switching Project at Washington University for the period from June 15, 1998 through September 15, 1998.

This paper presents a new communication service for ATM networks that provides one-way, adjustable rate, on-demand communication channels. The proposed dynamic flow service is designed to operate within a multi-service cell switched network that supports both conventional switched virtual circuits and IP packet routing and is designed to complement those services. It is particularly well suited to applications that transmit substantial amounts of data (a few tens of kilobytes or more) in fairly short time periods (up to a few seconds). Much of the current world-wide web traffic falls within... Read More

We introduce a model for learning from examples and membership queries in situations where the boundary between positive and negative examples is somewhat ill-defined. In our model, queries near the boundary of a target concept may receive incorrect or "don't care" responses, and the distribution of examples has zero probability mass on the boundary region. The motivation behind our model is that in many cases the boundary between positive and negative examples is complicated or "fuzzy." However, one may still hope to learn successfully, because the typical examples that one... Read More

We first present an algorithm that uses membership and equivalence queries to exactly identify a discretized geometric concept defined by the unioin of m axis-parallel boxes in d-dimensional discretized Euclidean space where each coordinate can have n discrete values. This algorithm receives at most md counterexamples and uses time and membership queries polynomial in m and log(n) for any constant d. Furthermore, all equivalence queries can be formulated as the union of O(mdlog(m)) axis-parallel boxes. Next, we show how to extend our algorithm to efficiently learn, from only equivalence queries,... Read More

We present several efficient parallel algorithms for PAC-learning geometric concepts in a constant-dimensional space. The algorithms are robust even against malicious classification noise of any rate less than 1/2. We first give an efficient noise-tolerant parallel algorithm to PAC-learn the class of geometric concepts defined by a polynomial number of (d-1)-dimensional hyperplanes against an arbitrary distribution where each hyperplane has a slope from a set of known slopes. We then describe how boosting techniques can be used so that our algorithms' dependence on {GREEK LETTER} and {DELTA} does not depend... Read More

We present an efficient algorithm for PAC-learning a very general class of geometric concepts over Rd for fixed d. More specifically, let T be any set of s halfspaces. Let x = (x1,...,xd) be an arbitrary point in Rd. With each t Є T we associate a boolean indicator function It(x) which is 1 if and only if x is in the halfspace t. The concept class Cds that we study consists of all concepts formed by any boolean function over It1, ...Its for ti Є T. This class is... Read More

Cluster based architectures that employ high performance inexpensive Personal Computers (PCs) interconnected by high speed commodity interconnect have been recognized as a cost-effective way of building high performance scalable Multimedia-On-Demand (MOD) storage servers [4, 5, 7, 9]. Typically, the PCs in these architectures run operating systems such as UNIX that have traditionally been optimized for interactive computing. They do not provide fast disk-to-network data paths and guaranteed CPU and storage access. This paper reports enhancements to the 4.4 BSD UNIX system carried out to rectify these limitations in the context... Read More

Current network technology is bandwidth-rich but latency-poor; thus round-trip delays will dominate access latency for web traffic. We describe four new techniques that reduce the round-trips needed for web accesses. The techniques are based on the paradigm of preprocessing a web page to collect information about links and inline data in the page. Stored Address Binding almost always eliminates the DNS lookup (which can cost seconds) at the start of a transaction. In Informed Server Proxying, a server tells its client that it has cached pages referenced in a page... Read More

This report develops a practical method for connection admission control in ATM networks. The method is based on a virtual cell loss probability criterion, is designed to handle heterogeneous traffic types and allows each traffic source to be described by an individual finite-state model with as many states as are needed to describe the source traffic. To make connection admission decisions with respect to individual links, an aggregate finite state model is computed from the individual models and used to estimate the virtual cell loss probabilities. To reduce the computational... Read More

We describe new mechanisms to deal with asymmetries that arise in routing protocols. We show how to avoid route asymmetries (due to non-unique shortest paths) by adding random integer link costs. We show in detail how RIP can be modified to avoid route asymmetry with high probability, without affecting either its efficiency or performance metrics such as convergence time. Symmetrical intra-domain routing also makes possible a new form of congestion control that we call Reverse Path Congestion Control (RPCC). We show, using simulations, that RPCC can augment existing TCP congestion... Read More

The Internet Inter-ORB Protocol (IIOP) enables heterogeneous CORBA-compliant Object Request Brokers (ORBs) to interoperate over TCP/IP networks. The IIOP uses the Common Data Representation (CDR) transfer syntax to map CORBA Interface Definition Langauge (IDL) data types into a bi-canonical wire format. Due to the excessive marshaling/demarshaling overhead, data copying, and high-levels of function call overhead, conventional implementation of IIOP protocols yield poor performance over high-speed networks. To meet the demands of emerging distributed multimedia applications, CORBA-compliant ORBs must support both interoperable and highly efficient IIOP implementations. This paper provides two... Read More

The objective of The Programmers' Playground, described in this manual, is to provide a development environment and underlying support for end-user construction of distributed multimedia applications from reusable self-describing software components. Playground provides a set of software tools and a methodology for simplifying the design and construction of applications that interact with each other and with people in a distributed computer system. This manual explains how to write interactive distributed applications using Playground. The only background necessary to get started is an understanding of basic data structures and control constructs... Read More

Developing the ability to recognize a landmark from a visual image of a robot's current location is a fundamental problem in robotics. We describe a way in which the landmark matching problem can be mapped to that of learning a one-dimensional geometric pattern. The first contribution of our work is an efficient noise-tolerant algorithm (designed using the statistical query model) to PAC-learn the class of one-dimensional geometric patterns. The second contribution of our work is an empirical study of our algorithm that provides at least some evidence that statistical query... Read More

The CORBA Event Service provides a flexible model for asynchronous communication among objects.However, the standard CORBA Event Service specification lacks important features required by real-time applications. For instance, operational flight programs for fighter aircraft have complex real-time processing requirements. This paper describes the design and performance of an object-oriented, real-time implementation of the CORBA Event Service that is designed to meet these requirements. This paper makes three contributions to the design and performance measurement of object-oriented real-time systems. First, it illustrates how to extend the CORBA Event Service so that... Read More

The non-determinism, complexity, and size of distributed software systems present significant difficulties for designers and maintainers. Visualization can help alleviate these difficulties through interactive exploratory tools that allow both novice and experienced users to investigate a distributed computation using a common tool set. Essential to the success of a visual exploration tool is the ability to provide accurate representations of global states. This paper is concerned with the use of snapshots in support of interactive visual exploration of distributed computations. The nature of the visualization process requires snapshots that (1)... Read More

High-performance Web servers are essential to meet the growing demands of the Internet. Satisfying these demands requires a thorough understanding of the key factors that affect Web server performance. This paper provides three contributions to the design, implementation, and evaluation of high-performance Web servers. First, we report the results of a comprehensive empirical study of popular high-performance Web servers (such as Apache, Netscape Enterprise, PHTTPD, and Zeus) over high-speed ATM networks. This study illustrates their relative performance and identifies their performance bottlenecks. To measure performance accurately, we developed a new... Read More

Computational steering, the interactive adjustment of application parameters and allocation of resources, is a promising technique for higher-productivity simulation, finer-grained optimization of dynamically varying algorithms, and greater understanding of program behavior and the characteristics of data sets and solution spaces. Tools for computational steering must provide monitoring, visualization, and interaction facilities. In addition, these tools must address issues related to the consistency, latency, and scalability at each of these phases, and must consider the perturbation that results. In this paper we describe transaction-based components for a computational steering system and... Read More

Formal accounts of negotiation tend to invoke the strategic models of conflict which have been impressively developed by game theorists in this half-century. For two decades, however, research on artificial intelligence (AI) has produced a different formal picture of the agent and of the rational deliberations of agents. AI's models are not based simply on intensities of preference and quantities of probability. AI's models consider that agents use language in various ways, that agents use and convey knowledge, that agents plan, search, focus, and argue. Agents can choose their language,... Read More

This paper estimates the link costs of constraint based nonblocking ATM networks on a sphere. Analytical results are obtained when switches are uniformly distributed on the surface of a unit sphere and every switch has source and sink capacity of one, and the results are compared with simulations.

Recent advances in wireless networking technology and the increasing demand for ubiquitous, mobile connectivity demonstrate the importance of providing reliable systems for managing reconfiguration and disconnection of components. Design of such systems requires tools and techniques appropriate to the task. Many formal models of computation, including UNITY, are not adequate for expressing reconfiguration and disconnection and are therefore inappropriate vehicles for investigating the impact of mobility on the construction of modular and composable systems. Algebraic formalisms such as the pi-calculus have been proposed for modeling mobility. This paper addresses the... Read More

Traditionally, a distributed system has been viewed as a collection of fixed computational elements connected by a static network. Prompted by recent advances in wireless communications rechnology, the emerging field of mobile computing is challenging these assumptions by providing mobile hosts with connectivity that may change over time, raising the possibility that hosts may be called upon to operate while only weakly connected to or while completely disconnected from other hosts. We define a concurrent mobile system as one where independently executing coponents may migrate through some space during the... Read More

Aggregate visualization and manipulation enables the viewing and interaction of dynamically changing data sets in a graphically meaningful way. However, off-the-shelf applications generally provide only limited ways to view aggregates. To be truly effective to the end-user, an aggregate visualization should be customizable to suit the individual's needs. This paper describes a software system that empowers end-users to create interactive aggregate visualizations through direct manipulation. Included are mechanisms for specifying how aggregate data is processed from multiple sources, providing functionality similar to project, select, join, and cross product of relational... Read More

Aggregate visualization and manipulation enables the viewing and interaction of dynamically changing data sets in a graphically meaningful way. However, off-the-shelf applications typically provide only limited ways to view static aggregates and generally to not support manipulation of aggregate data through the resulting visualization. To be fully dynamic, an aggregate visualization should be customizable to suit the individual's needs and should allow end-users to modify the data through direct manipulation. This paper describes a software system that empowers end-users to create interactive aggregate visualizations through a visual language interface. Included... Read More

EUPHORIA is a user interface management system that enables end-users to create direct manipulation graphical user interfaces (GUIs) through interactive drawing. Used in conjunction with The Programmers' Playground, a distributed programming environment, end-users can dynamically create and associate GUI components with an underlying application without programming, This document describes EUPHORIA's functionality.

With recent advances in wireless communication and the ubiquity of laptops, mobile computing has become an important research area. An essential problem in mobile computing is the delivery of a message from a source to either a single mobile node, unicast, or to a group of mobile nodes, multicast. Standard solutions proposed for macromobility (Mobile IP) and micromobility (cellular phones) for the unicast problem rely on tracking the mobile node. Tracking solutions scale badly when mobile nodes move frequently, and do not generalize well to multicast delivery. Our paper proposes... Read More

Retransmission based error control for large scale multicast applications is difficult because of two main problems: request implosion and lack of local recovery. Existing schemes (SRM, RMTP, TMTP, LBRRM) have good solutions to request implosion, but only approximate solutions (e.g., based on scoped multicast) for the local recovery problem. Our scheme achieves finer grain fault recovery by exploiting new forwarding services that allow us to create a dynamic hierarchy of receivers. We use a new paradigm, where routers provide a more refined form of multicasting (that may be useful to... Read More

We propose a network monitoring, visualization and control system (NMVC) that ensures adequate quality of service to network users while maintaining high network resource utilization. The main components of our system are a network probe, an endsystem probe, software network management agents that provide extensible multi-attribute event filtering for highly scalable data/event collection, network operation centers (NOCs) which can remotely install and (re)configure these agents, efficient online event ordering algorithms that can help synthesize and display a consistent view of network health, status and performance and a View Choreographer that... Read More

Advancements in network technology have led to the emergence of new computing paradigms that challenge established programming practices by employing weak forms of consistency and dynamic forms of binding. Code mobility, for instance, allows for invocation-time binding between a code fragment and the location where it executes. Similarly, mobile computing allows hosts (and the software they execute) to alter their physical location. Despite apparent similarities, the two paradigms are distinct in their treatment of location and movement. This paper seeks to uncover a common foundation for the two paradigms by... Read More

Advancements in network technology have led to the emergence of new computing paradigms that challenge established programming practices by employing weak forms of consistency and dynamic forms of binding. Code mobility, for instance, allows for invocation-time binding between a code fragment and the location where it executes. Similarly, mobile computing allows hosts (and the software they execute) to alter their physical location. Despite apparent similarities, the two paradigms are distinct in their treatment of location and movement. This paper seeks to uncover a common foundation for the two paradigms by... Read More

Advocates of software design patterns claim that using design patterns improves communication between software developers. The controled experiment that we describe in this report tests the hypothesis that software maintainers of well-structured, well-documented software containing design patterns can make changes (1) faster and (2) with less errors if the use of patterns is explicitly documented in the software. The experiment was performed with 22 participants of a university course on C++ and design patterns; it is similar to a previous experiment performed in Karlsruhe. For one of the two experiment... Read More

Traditionally, a distributed system has been viewed as a collection of fixed computational elements connected by a static network. Prompted by recent advances in wireless communications rechnology, the emerging field of mobile computing is challenging these assumptions by providing mobile hosts with connectivity that may change over time, raising the possibility that hosts may be called upon to operate while only weakly connected to or while completely disconnected from other hosts. We define a concurrent mobile system as one where independently executing coponents may migrate through some space during the... Read More

Regions of DNA rich in CpG dinucleotides, also known as CpG islands, are often located upstream of the transcription start side in both tissue specific and housekeeping genes. Overall, CPG dinucleotides are observed at a density of 25% the expected level from base composition alone, partially due to 5-methylcytosine decay (Bird, 1993). Since CpG dinucleotides typically occur with low frequency, CpG islands can be distinguished statistically in the genome. Our method of detecting CpG islands involves a heuristic algorithm employing classic changepoint methods and log-likelihood statistics. A Java applet has... Read More

DNA sequence analysis depends on the accurate assembly of fragment reads for the determination of a consensus sequence. Genomic sequences frequently contain repeat elements that may confound the fragment assembly process, and errors in fragment assembly, and errors in fragment assembly may seriously impact the biological interpretation of the sequence data. Validating the fidelity of sequence assembly by experimental means is desirable. This report examines the use of restriction digest analysis as a method for testing the fidelity of sequence assembly. Restriction digest fingerprint matching is an established technology for... Read More

Application Management permits the advertising, launching, and configuring of distributed applications created using the Programmers' Playground. Applications can be documented and made available to end-users through the use of application pages on the World Wide Web. The launching and configuring of applications is performed by a brokerage system consisting of an applicatoin broker and one or more hierarchies of module launchers. This document describes how to setup and use the components of the Application Management system.

Cappuccino is a planning tool for topological design of ATM networks. It uses a novel constraint-based approach to ATM network design. Extensibility of the tool is a basic design goal and the tool provides an open interface to incorporate new algorithms.

This report summarizes the results of an architectural study on Terabit Burst Switching. The purpose of this study was to explore alternative architectures for very high performance switching for data communication, using a combination of optical and electronic technologies. We explore two alternative implementations of the burst switching concept in detail, one using a hybrid architecture with an electronic core, and an integrated architecture using an all optical data path. We also briefly discuss an approach using optical TDM. Our results show that using the hybrid architecture, it is feasible... Read More

The rapid development of Asynchronous Transfer Mode technology in the last 10-15 years has stimulated renewed interest in the design and analysis of switching systems, leading to new ideas for system designs and new insights into the performance and evaluation of such systems. As ATM moves closer to realizing the vision of ubiquitous broadband ISDN services, the design of switching systems takes on growing importance. This paper seeks to clarify the key architectural issues for ATM switching system design and provides a survey of the current state-of-the-art.

In a previous paper [3], one of the authors, gave a worst-case analysis for the Early Packet Discard (EPD) technique for maintaining packet integrity during overload in ATM switches. This analysis showed that to ensure 100% goodput during overload under worst-case conditions, requires a buffer with enough storage for one maximum length packet from every active virtual circuit. This paper refines that analysis, using assumptions that are closer to what we expect to see in practice and examines how EPD performs when the buffer is not large enough to achieve... Read More

We reconsider several issues related to fragmentation and reassembly in IP. We first reconsider reassembly. We describe a simple expected case optimization that improves reassembly performance to 38 instructions per fragment if the fragments arrive in FIFO order (the same assumption made in header prediction) which has been implemented in the NetBSD kernel. Next, we introduce the new idea of Graceful Intermediate Reassembly (GIR), which is a generalization of the existing IP mechanisms of destination and hop-by-hop reassembly. In GIR, we coalesce the fragments at an intermediate router in order... Read More

This report describes the design and implementation of a gigabit ATM switching system supporting link rates from 150 Mb/s to 2.4 Gb/s, with a uniquely efficient multicast switch architecture that enables the construction of systems with essentially constant per port costs for configurations ranging from 8 to 4096 ports and system capacities approaching 1- Tb.s. The system design supports many-to-one and many-to-many forms of multicast, in addition to the usual one-to-many. It also provides multicast virtual paths, constant time configuration of multicast connections and an efficient packet-level discard method, that... Read More

This paper presents the ARAS simulator with which asynchronous instruction pipelines can be modelled, simulated and displayed. ARAS allows one to construct instruction pipelines by preparing various configuration files. Using these files and a number of benchmark programs, performance of the instruction pipelines can be obtained. The performance of asynchronous instruction pipelines can also be compared to synchronous case. Thus, one can decide the optimal design for instruction pipelines in asynchornous or synchronous cases and explore the deisng space of asynchronous instruction pipeline architectures.

This paper outlines a study of MPEG compressed video sequences and simulation of multiplexed video traffic in the ATM environment. A number of statistical characteristics including autocorrelation and variance of MPEG-1 compressed video sequences are used to characterize the 16 sample traces used in this study. From these measurements, a preliminary model is developed which utilizes basic measurements of the individual component video sequences to predict bandwidth requirements and cell loss of the multiplexed video traffic.

We present the design and implementation of Sensus, a practical, secure and private system for conducting surveys and elections over computer networks. Expanding on the work of Fujioka, Okamoto, and Ohta, Sensus uses blind signatures to ensure that only registered voters can vote and that each registered voter only votes once, while at the same time maintaining voters' privacy. Sensus allows voters to verify independently that their votes were counted correctly, and anonymously challenge the results should their votes be miscounted. We outline seven desirable properties of voting systems and... Read More

We are building a very high performance 1.2 Gb/s ATM network interface chip called the APIC (ATM Port Interconnect Controller). In addition to borrowing userful ideas from a number of research and commercial prototypes, the APIC design embraces several innovative features, and integrates all of these pieces into a coherent whole. This paper describes some of the novel ideas that have been incorporated in the APIC design with a view to improving the bandwidth and latency seen by end-applications. Among the techniques described, Protected DMA and Protected I/O were designed... Read More

This paper considers the problem of designing ATM networks that are nonblocking with respect to virtual circuit requests, subject to specified constraints on the traffic. In this paper, we focus on global traffic constraints that simply limit the total entering and exiting traffic at each switching system. After reviewing prior results for linear link costs, we introduce a more realistic link cost model, and develop a number of results using it. We also describe a technique for converting tree-structured networks to nonblocking hierarchical networks satisfying limits on the capacity of... Read More

This paper addresses the problem of topological design of ATM (and similar) communication networks. We formulate the problem from a worst-case point of view, seeking network desings that, subject to specified traffic constraints, are nonblocking for point-to-point and multicast virtual circuits. Within this model we give various conditions under which star networks are optimal or near-optimal. These conditions are approximately satisfied in many common situations making the results of practical significance. An important consequence of these results is that, where they apply, there is no added cost for nonblocking multicast... Read More

This paper seeks to bridge the gap between theory and practice of real-time scheduling in the domain of multimedia computer systems. We show that scheduling algorithms that are good in theory, often have practical limitations. However when these algorithms are modified based on practical considerations, existing theoretical results cannot be used as they are. In this paper we motivate the need for new scheduling schemes for multimedia protocol processing, and demonstrate their real-time performance in our prototype implementation. We then explain the observed results by analysis and measurement. More specifically,... Read More

Real-time upcalls (RTUs) are an operating systems mechanism to provide quality-of-service (QoS) guarantees to network applications, and to efficiently implement protocols in user space with (QoS) guarantees. Traditionally, threads (and real-time extensions to threads) have been used to structure concurrent activities in user space protocol implementations. However, preemptive scheduling required for real-time threads leads to excessive context switching, and introduces the need for expensive concurrency control mechanisms such as locking. The RTU mechanism exploits the iterative nature of protocol processing to eliminate the need for locking, and reduce asynchronous preemption,... Read More

We report a number of new results in generalized caching. This problem arises in modern computer networks in which data objects of various sizes are transmitted frequently. First it is shown that its optimal solution is NP-complete. Then we explore two methods of obtaining nearly optimal answers based on the dynamic programming algorithm that we provided in [5]. These methods enable a trade-off between optimality and speed. It is also shown that LFD (the longest forward distance algorithm which is the optimal policy in the classical case), is no longer... Read More

Network traffic can be reduced significantly if caching is utilized effectively. As an effort in this direction we study the replacement problem that arises in caching of multimedia objects. The size of objects and the cost of cache misses are assumed non-uniform. The non-uniformity of size is inherent in multimedia objects, and the non-uniformity of cost is due to the non-uniformity of size and the fact that the objects are scattered throughout the network. Although a special case of this problem, i.e. the case of uniform size and cost, has... Read More

This report describes an ATM Multipoint Connection Caching strategy (AMCC) to control the explosive growth of traffic within the network and at an endpoint in a large Distributed Interactive Simulation (DIS) application such as a battlefield simulation. For very large DIS applications with 100,000 entities, the current method of broadcasting information among entities will no longer be feasible due to computational and network bandwidth limitations. Our scheme divides the simulation space into grids and each grid square or a set of grid squares forms a multicast group. Entities join the... Read More

With recent advances in wireless communication technology, mobile computing is an increasingly important area of research. A mobile system is one where independently executing components may migrate through some space during the course of the computation, and where the pattern of connectivity among the components changes as they move in and out of proximity. Mobile UNITY is a language and logic for specifying and reasoning about mobile systems, the components of which must operate in a highly decoupled way. In this paper it is argued that Mobile UNITY contributes to... Read More

Distributed multimedia applications supported by a global electronic infrastructure have tremendous potential for providing users with customized communication and computation environments. Since communication and computation requirements vary by context and change dynamically, it is unlikely that off-the-shelf applications will anticipate the needs of all users. Therefore, empowering end-users to create their own customized applications for both communication and computation is an important challenge. This dissertation presents several mechanisms that enable end-users to create and configure distributed multimedia applications, including end-users construction direct manipulation graphical users interface (GUIs) and application management... Read More

This paper describes an empirical study of end-users that tested the usability of The Programmers' Playground graphical environment. The Programmers' Playground is a software library and run-time system for constructing distributed multimedia applications. Playground's graphical environment enables end-users to create direct manipulation graphical user interfaces (GUIs) and to dynamically configure communication among distributed application components. In this study, 28 end-users with no prior experience in distributed computing or user interface construction were timed and evaluated on several tasks using our graphical environment. Tasks included the use of direct and indirect... Read More

With recent advances in wireless communication and the ubiquity of laptops, mobile computing has become an important research area. An essential problem in mobile computing is the delivery of a message from a source to either a single mobile node, unicast, or to a group of mobile nodes, multicast. Standard solutions used in Mobile IP and cellular phones for the unicast problem rely on tracking the mobile unit. Tracking solutions scale badly when mobile nodes move frequently, and do not generalize well to multicast delivery. Our paper proposes a new... Read More

We present a voice-activated, hands-off, ATM-based video conferencing application. The application, called Vaudeville, features high quality NTSC video, voice-activated audio transmission, audio bridging of two audio streams, and voice-activated video switching. It supports multiple simultaneous multi-party conferences using a scalable multicast mechanism. We describe how Vaudeville was built using a component-based distributed programming environment. We also describe the algorithms used to contorl the audio and video of the applciation. Audio and video are encoded in hardware using an ATM hardware multimedia interface.

Software developers typically must choose between interpreted and compiled environments for their programming activities. However, the current trends toward mobile computing and platform independence suggest moving to a new continuous compilation paradigm that integrates the advantages of each environment. Movement in this direction can already be seen in the development of Sun Microsystems' Java environment. The resulting continuous compiler operates not as a prelude to, but rather in tandem with, program execution. In this thesis we present the results of experiments that compare the performance of the continuous compilation model... Read More

Design patterns are a topic of great current interest within the object-oriented programming community. The motivation is both economical and intellectual. On one hand, there is the hope of establishing a common culture and language that fosters communicatino and growth in the software engineering field. While a community dominated by empiricism is seeking to achieve higher levels of formality by capturing its experiences in the form of catalogs of design patterns, another community, deeply rooted in formal thinking, is seeking to make its mark on the every day workings of... Read More

Mobile computing represents a major point of departure from the traditional distributed computing paradigm. The potentially very large number of independent computing units, a decoupled computing style, frequent disconnections, continuous position changes, and the location-dependent nature of the behavior and communication patterns present designers with unprecedented challenges in the areas of modularity and dependability. So far, the literature on mobile computing is dominated by concerns having to do with the development of protocols and services. This paper complements this perspective by considering the nature of the underlying formal models that... Read More

As computer size continues to decrease and new user interface technologies become more ubiquitous, the conventional keyboard and mouse input interfaces are becoming harder to design into newer machines and less practical for use in some applications. The pen is one input technology more suited for the upcoming generation of smaller computers using direct manipulation interfaces. However, a pen-only user interface relies on continuous gesture and handwriting tecognizers that are often slow, inaccurate, and error prone for command and text entry. Speech recognition is an input modality that can input... Read More

Distributed stream filtering is a mechanism for implementing a new class of real-time applications with distributed processing requirements. These applications require scalable architectures to support the efficient processing and multiplexing of large volumes of continuously generated data. This paper provides an overview of a stream-oriented model for database query processing and presents a supporting implementation. To facilitate distributed stream filtering, we introduce several new query processing operations, including pipelined filtering that efficiently joins and eliminates duplicates from database streams and a new join method, the progressive join, that joins streams... Read More

We describe an efficient fair queuing scheme, Leap Forward Virtual Clock, that provides end-to-end delay bounds almost identical to that of PGPS fair queuing, along with throughput fairness. Our scheme can be implemented with a worst-case time O(loglogN) per packet guaranteed delay and throughput fairness. As its name suggests, our scheme is based on Zhang's virtual clock. While the original virtual clock scheme does not achieve throughput fairness, we can modify it with a simple leap forward mechanism that keeps the server clock from lagging too far behind the packet... Read More

The main economic-theoretic approcahes to the problem of resource allocation make little if any reference to negotiation processes. These processes are fundamentally linguistic, based on the exchange of messages among agents. Communication being so fundamental in the characterization of negotiation processes, the analysis of negotiation must emphasize on the structure of the language in which the negotiations take place. Computer science and particularly Artificial Intelligence have provided interesting insights about that linguistic structure. In the first part of this work we present a brief survey of the literature on resource... Read More

One of the important features of ATM networks is their ability to support multicast communications. This facilitates the efficient distribution of multimedia information streams (such as audio and video) to large groups of receivers (potentially millions). Because ATM networks do not provide reliable delivery mechanisms, it is up to end systems to provide end-to-end reliability where it is needed. While this is straightforward for point-to-point virtual circuits, it is more difficult for one-to many and many-to-mamy virtual circuits. In this report, we propose some minimal extensions to the hardware of... Read More

Link striping algorithms are often used to overcome transmission bottlenecks in computer networks. However, traidtional striping algorithms suffer from two major disadvantages. They provide inadequate load sharing in the presence of variable length packets, and may result in non-FIFO delivery of data. We describe a new family of link striping algorithms that solve both problems. Our scheme applies to packets at any layer (physical, data, link, network, and transport) that work over multiple FIFO channels. We deal with variable sized packets by showing how a class of fair queueing algorithms... Read More

Distributed Radiological Multimedia Conference (DRMC) is a collaborative imaging/multimedia conferencing tool which allows geographically separated physicians to confer over a shared projection radiograph. DRMC utilizes the advantages of high bandwidth and scalability offered by the new Asynchronous Transfer Mode (ATM) network technology. This application is customized for the high quality of displayed images and rapid response to user requests. It allows conferees to: share a common radiograph; each possess an independently controlled globally visible cursor; be able to point to and outline areas on the image to bring it to... Read More

Synchronized Data Objects (SDOs) are presented as a method of encapsulating, in the datatype definition, synchronization protocols that are used to control information exchange. SDOs are presented in the context of I/O abstraction, a programming model that seeks to separate communication from computation in order to support dynamic end-user configuration of distrivuted applications. SDOs can be used to implement a variety of synchronization paradigms, including remote invalidation, demand-driven data streams, remote procedure call, and promises. An implementation of SDOs is described in the context of The Programmers' Playground, a distributed... Read More

Large scale storage servers that provide location transparent, interactive access to hundreds or thousands of concurrent, independent clients will be important components of hte furture information super-highway infrastructure. Two key requirements of such servers are as follows: support high parallelism and concurrency in data access to allow large number of access to the same or different data. Second, support independent interactive playout control operations such as fast-forward, rewind, slow-play, pause, resume, random access etc. with minimal latency. This paper assumes a distributed storage server architecture consisting of several high performance... Read More

In this paper, an asynchronous pipeline instruction simulator, ARAS is presented. With this simulator, one can design selected instruction pipelines and check their performance. Performance measurements of the pipeline configuration are obtained by simulating the execution of benchmark programs on the machine architectures developed. Depending on the simulation results obtained by using ARAS, the pipeline configuration can be altered to improve its performance. Thus, one can explore the design space of aynchronous pipeline architectures.

We describe a new implementation of the BSD callout and timer facilities. Current BSD kernels take time proportional to the number of outstanding timers to set or cancel timers. Our implementation takes constant time to start, stop, and maintain timers; this leads to a highly scalable design that can support thousands of outstanding timers without much overhead. Unlike the existing implementation, our routines are guaranteed to lock out interrupts only for a small, bounded amount of time. We also extend the setitimer() interface to allow a process to have multiple... Read More

A useful way to design simple and robust protocols is to make them self-stabilitizing. We describe a new general technique for self-stabilization called window washing. We apply this technique to generalized sliding window protocols that work on a number of topologies. This results in simple, efficient, and self-stabilizing protocols. As far as we know, both window washing and generalized sliding window protocols are new ideas. Our protocols can be used for data links, reliable broadcast, and flow control.

The goal of this research is to determine whether declared strategy voting can be an effective tool for group decision-making. Declared strategy voting is a novel group decision-making procedure in which preference is specified using voting strategies - first-order mathematical functions that specify a choice in terms of zero or more parameters. This research will focus on refining the declared strategy voting concept, developing an accessible implementation of declared strategy voting that can be used for mock elections, assessing the potential impacts of declared strategy voting, and evaluating the effectiveness... Read More

Effective load balancing algorithms are crucial in fully realizing the performance potential of parallel computer systems. This paper proposes a general matrix iterative model to represent a range of dynamic load balancing algorithms. The model and associated performance measures are used to evaluate and compare vairous load balancing algorithms and derive optimal algorithms and associated parameters for a given application and multiprocessor system. The model is parameterized to represent three load balancing algorithms - the random strategy, diffusion and complete redistribution algorithms. The model is validated by comparing the results... Read More

We present a new tool for automatically generating prototypes of communication protocols on a wide variety of platforms. Our goal is to reduce design time, enhance portability, and accommodate optimizations automatically. Users of the tool are required to provide an abstract implementation of the protocol in C++ without worrying about the underlying operating system specific system calls. Instead, the user employs high-level interface functions provided by the tool to interact with the underlying operating system. Users also need not worry about complex packet formats that involve fields of various bit... Read More

Developing the ability to recognize a landmark from a visual image of a robot's current location is a fundamental problem in robotics. We consider the problem of PAC-learning the concept class of geometric patterns where the target geometric pattern is a configuration of k points on the real line. Each instance is a configuration of n points on the real line, where it is labeled according to whether or not it visually resembles the target pattern. To capture the notion of visual resemblance we use the Hausdorff metric. Informally, two... Read More

Real-time upcalls (RTUs) are an operating systems mechanism that can be used by applications to efficiently schedule code segments (or handlers) that must execute periodically. While the mechanism was conceibed to support protocol processing with quality-of-service guarantees for networked multimedia applicatoins it is general enough to be applicable in other domains like real-time image processing. Until now real-time threads have been the only mechanism for implementing protocols in user space with QoS guarantees. The RTU mechanism avoids the implementation complexity of the thread based approach while retaining its ability to... Read More

Packet filters are a mechanism for efficiently demultiplexing network packets to application endpoints. There is currently no general, formal specification method for packet filters that allows for easy or efficient composition of specifications. In this paper we present an automatic approach that achieves all of these goals. We approach packet filter specification as a language recognition problem: each filter is represented by a context-free grammar, whose language is the set of packets the filter should accept. Thus, packet filters can be formulated through a general, well defined specification; further, the... Read More

This thesis presents a simple, yet powerful, set of mechanisms for testing and debugging distributed applications consisting of modules that communicate through well-defined data interfaces. The tools allow default or programmer-defined functions to be attached to various communication events so that particular data values at interesting points in the program are made available for testing and debugging. The debugging status of each component of the communication interface can be controlled separately so that various debugging information can be turned on and off during program execution. By attaching breakpoints to programmer-defined... Read More

Hart's "Ascription of Responsibility and Rights" is where we find perhaps the first clear pronouncement of defeasibility and the technical introduction of the term. The paper has been criticised, disavowed, and never quite fully redeemed. Its lurid history is now being used as an excuse for dismissing the importance of defeasibility. Quite to the contrary, Hart's introduction of defeasibility has uniformly been regarded as the most agreeable part of the paper. The critics' wish that defeasibility could be better expounded along the lines of a Wittgensteinian game-theoretic semantics has largely... Read More

Previous teaching models in the learning theory community have been batch models. That is, in these models the teacher has generated a single set of helpful examples to present to the learner. In this paper we present an interactive model in which the learner has the ability to ask queries as in the query learning model of Angluin [1]. We show that this model is at least as powerful as previous teaching models. We also show that anything learnable with queries, even by a randomized learner, is teachable in our... Read More

Constraints are widely recognized as a useful tool for user interface constructino. Through constraints, relationships among user interface components can be defined declaratively, leaving the task of relationship management to a constraint solver. Multi-way constraint solvers supporting constraint hierarchies provide a means to specify preferential constraint relationships with a dynamically changing computation flow, making them especially well suited to interactive user interfaces. However, previous such constraint solvers lack the ability to enforce inequalities or to effectively handle cyclic constraint relationships. These deficiencies limit the problems that could be solved using... Read More

The Programmers' Playground is a software library and run-time system for creating distributed multimedia applications from collections of reusable software moduels. This paper presents the design and implementation of EUPHORIA, Playground's user interface management system. Implemented as a Playground module, EUPHORIA allows end-users to create direct manipulation graphical user interfaces (GUIs) exclusively through the use of a graphics editor. No programming is required. At run-time, attributes of the GUI state can be exposed and connected to external Playground modules, allowing the user to vosualize and directly manipulate state information in... Read More

High-bandwidth multimedia applications pose new challenges to error control. These include the support of error control for Continuous Media (CM) streams and the scalable support of error control in multipoint applications where the number of participants is large. Current error control mechanisms provide no support for the above applications. In this report we present new error control mechanisms that provide the required support. Continuous media applications have strict timing requirements which greatly affect recovery. To support continuous media applications we have designed and implemented a point-to-point error control mechanism which... Read More

Mobile computing represents a major point of departure from the traditional distributed computing paradigm. The potentially very large number of independent computing units, a decoupled computing style, frequent disconnections, continuous position changes, and the location-dependent nature of the behavior and communication patterns present designers with unprecedented challenges in the areas of modularity and dependability. This paper describes a modular approach to specifying and reasoning about of mobile computing. Its novelty rests with the notion of allowing transient (location-dependent) data sharing among programs which move in space. The notation is a... Read More

Mobile computing is emerging as an important new paradigm which has the potential to reshape our thinking about distributed computation. Mobility has far-reaching implications on what designers and users can assume about communication patterns, resource availability, and applciation behaviors as components move from one location to another while joining or leaving groups of other components in their vicinity. New distributed algorithms are likely to be required as the nature of applications shifts with the emergence of this new kind of computing environment. Formal methods have an important role to play... Read More

Mobile computing represents a major point of departure from the traditional distributed computing paradigm. The potentially very large number of independent computing units, a decoupled computing style, frequent disconnections, continuous position changes, and the location-dependent nature of the behavior and communication patterns of the individual components present designers with unprecedented challenges in the areas of modularity and dependability. This paper describes two ideas regarding a modular approach to specifying and reasoning about mobile computing. The novelty of our approach rests with the notion of allowing transient interactions among programs which... Read More

This paper describes the design of the ACE object-oriented thread encapsulation class library. The architecture of this class library is presented from an end-user and internal design perspective and several key design issues are discussed. Readers should gain an understanding of the overall design approach as well as the tradeoffs between various software quality factors such as performance, portability, and extensibility.

The algorithmic analysis, data collection, and statistical analysis required to isolate the cause of time variability observed while an Elman style recurrent neural network is trained in parallel on a twenty processor SPARCcenter 2000 is described in detail. Correlations of system metrics indicate the operating system scheduler or an interaction of kernel processes is the most probable explanation for the variability.

The Programmers' Playground is a programming environment that supports end-user construction of distributed multimedia applications. The system implements a new programming model that is based, in part, upon ideas from the formal I/O automaton model of Lynch and Tuttle. Important features of The Programmers' Playground are a separation of communication and computation and graphical support for dynamic reconfiguration. This paper provides a formal specification of the Playground programming model and runtime system in terms of the I/O automaton model on which it is based. Exploiting the compositionality properties of the... Read More

Asynchronous Transfer Mode (ATM) network technology is expected to become a central part of the emerging global information infrastructure. ATM networks introduce a number of features that distinguish them from earlier technologies and introduce new issues in network control. This paper offers a framework for precisely defining and analyzing alternative approaches to the distributed control of ATM networks and explores some of the key design issues through a series of examples. It is hoped that it will provide a useful foundation for researchers in networking and distributed computing interested in... Read More

This report analyzes two popular heuristics for ensuring packet integrity in ATM switching systems. In particular, we analyze the behavior of packet tail discarding, in order to understand how the packet level link efficiency is dependent on the rates of individual virtual circuits and the degre of the imposed overload. In addition, we study early packet discard and show that the queue capacity needed to achieve high efficiency under worst-case conditions grows with the number of virtual circuits and we determine the efficiency obtainable with more limited queue capacities. Using... Read More

As ATM becomes widely accepted as the communication standard for high speed networks, the signaling system structure and protocols that support ATM become more and more important. To support existing, future and unknown applications, the signalign system has to be very flexible and efficient. In this paper we define the signaling problem, present several possible signaling system structures, compare the advantages and disadvantages of these systems, and then we propose a new signaling system structure. The fundamental idea of the new signaling system is the logical separation of the signaling... Read More

Abstract Network signaling is the process of transferring control information among components of a communication network to establish, maintain, and release connections, and to pass the network management information. The rapid evolution in the field of telecommunications has led to the rapid evolution of network signaling. In this paper, we review the evolution of network signaling. We emphasize the concepts and protocols used in modern fast packet switching networks especially in emerging ATM networks.