Program Information

Please note that the content of this web page is only a subset of the program information available in the downloadable Printed Advance Program Information. The Final Program Information is available as of Wednesday, 12th of June.

Welcome Reception

ARC Seibersdorf research GmbH is the sponsor of our welcome reception to be held on Monday evening (6:30-7:30 pm) at the Parkhotel Schönbrunn's "Sissi-Salon". The welcome reception is the ideal place to socialize with other participants before the conference! It will also be possible to pick up the registration material (badge, proceedings, ...) at the welcome reception, allowing participants to skip queues on Tuesday morning.

Printed Program

The printed program is available as of Friday, April 26th. To get a copy of the printed program, send your postal address to the conference secretariat.

If you want to receive multiple copies for local redistribution, please include how many you need together with an indication of where you intend to distribute them. Thanks for any help.

Invited Speakers

Embedded Systems Unsuitable for Object Orientation
Maarten Boasson
Quaerendo Invenietis bv & University of Amsterdam

Abstract
It will be argued that the current focus on object technology is detrimental to progress in embedded systems. The core of the problem is that OO is fine for analysis but does not answer the design needs. Solutions for shortcomings are sought within the OO dogma, making things worse. This talk will outline a different approach.

Short biography

Maarten Boasson studied mathematics in Groningen, the Netherlands. He started his career with Hollandse Signaalapparaten, a defense contractor specializing in naval systems. He became involved in advanced studies aiming at control of complexity, both of the development process and of the system under development itself. This resulted in the creation of a novel architecture for distributed reactive systems, that has been applied successfully in numerous systems and is, more than 10 years after its introduction, still unsurpassed in its support for integration, fault tolerance and component reuse, in particular in combination with high performance. In 1996 Boasson was appointed professor of computer science at the University of Amsterdam, where he holds a chair in Industrial Complex Computer Systems. He has been involved in the organisation of numerous international conferences, played a major role in establishing a dutch national research programme in embedded systems, and is currently associate editor-in-chief of IEEE Software.

Contextware: Bridging Physical and Virtual Worlds
Alois Ferscha
University of Linz

Abstract
A variety of terms - like Ubiquitous Computing, Pervasive Computing, Invisible Computing, Ambient Intelligence, Sentient Computing, Post-PC Computing etc. - today refers to new challenges and paradigms to the interaction between users and mobile and embedded computing devices. Fertilized by a vast quantitative growth of the Internet over the past years and a growing availability of wireless communication technologies in the wide, local and personal area, an ubiquitous use of ``embedded'' information society technologies is evolving. Most of the services delivered through those new technologies are services adapted to context, particularly to the person, time and location of their use.

This talk will explore the software engineering issues, challenges and enabling technologies associated with the provision of context-aware services. In analogy to the term ``middleware'' - generally understood as software technologies that serve to mediate between two or more separate and usually already existing software components - the term ``contextware'' is introduced as the core of software technologies mediating services and the context of their use, thus bridging virtual and physical worlds.

Short biography

Alois Ferscha received the Mag. degree in 1984, and a PhD in business informatics in 1990, both from the University of Vienna, Austria. From 1986 through 2000 he was with the Department of Applied Computer Science at the University of Vienna at the levels of assistant and associate professor. In 2000 he joined the University of Linz as full professor. Prof. Ferscha has published more than 60 technical papers on topics related to parallel and distributed computing, like e.g. Computer Aided Parallel Software Engineering, Performance Oriented Distributed/Parallel Program Development, Parallel and Distributed Discrete Event Simulation, Performance Modeling/Analysis of Parallel Systems and Parallel Visual Programming. Currently he is focussed on Pervasive Computing, Embedded Software Systems, Wireless Communication, Multiuser Cooperation, Distributed Interaction and Distributed Interactive Simulation.

He has been the project leader of several national and international research projects like e.g.: Network Computing, Performance Analysis of Parallel Systems and their Workload, Parallel Simulation of Very Large Office Workflow Models, Distributed Simulation on High Performance Parallel Computer Architectures, Modelling and Analysis of Time Constrained and Hierarchical Systems (MATCH, HCM), Broadband Integrated Satellite Network Traffic Evaluation (BISANTE, ESPRIT IV) and Distributed Cooperative Environments (COOPERATE) and Virtual Enterprises. Currently he is pursuing project work related to context based application frameworks in a "Wireless Campus" network, public communication displays with wireless remote controls, geo-enhanced mobile navigation systems, RFID based realtime notification systems, wearable computing and embedded internet application frameworks.

He has been a visiting researcher at the Dipartimento di Informatica, Universita di Torino, Italy, at the Dipartimento di Informatica, Universita di Genoa, Italy, at the Computer Science Department, University of Maryland at College Park, College Park, Maryland, U.S.A., and at the Department of Computer and Information Sciences, University of Oregon, Eugene, Oregon, U.S.A. He has served on the committees of several conferences like WWW, PADS, DIS-RT, SIGMETRICS, MASCOTS, TOOLS, PNPM, ICS, and was programe committee chair of PADS'98, MASCOTS'99, co-chair of the Pervasive Computing Workshop at Informatik 2001 and vice-chair of WWW2002. Prof. Ferscha is member of the GI, ACM and IEEE and holds the Heinz-Zemanek Award for distinguished contributions in computer science.

Indulgent Algorithms: Virtues and Limitations
Rachid Guerraoui
Swiss Federal Institute of Technology in Lausanne (EPFL)

Abstract
An indulgent algorithm is a reliable distributed algorithm that does not need reliable failure detection. In particular, with an indulgent algorithm, no process ever needs to know if some other process is up or down. This talk will discuss some practical and theoretical ramifications of indulgent algorithms, and will give some new results on indulgent algorithms in the context of the consensus problem.

Short biography

Rachid Guerraoui is professor in computer science at the Swiss Federal Institute of Technology in Lausanne where he leads the Distributed Programming Laboratory. He is interested in classic and less classic music, triathlons, object-oriented programming and distributed algorithms.

On Architectural Stability and Evolution
Mehdi Jazayeri
Technical University of Vienna

Abstract
Many organizations are now pursuing software architecture as a way to control their software development and evolution challenge. A software architecture describes the properties of a family of products, thus addressing the problems of both development and evolution. An important problem is to be able to evaluate the "goodness" of a proposed architecture. In this talk, I propose stability or resilience as a measure of goodness of an architecture. The stability of an architecture is a measure of how well it accommodates new family members, and we can measure that by the amount of code changes necessary for the introduction of a new member. I review the results of a case study of several releases of a telecommunication software system containing a few million lines of code to demonstrate one way to try to estimate architectural stability. I will review the challenges in software evolution and conclude with recommendations for future research. This work was part of the project Architectural Reasoning for Embedded Systems (ARES).

Keywords:
software releases, release histories, software architecture, product family, architectural stability, software visualization

Short biography

Mehdi Jazayeri is a professor of computer science at the Technical University of Vienna where he holds the chair of distributed systems. He spent many years in software research and development at several Silicon Valley companies, including ten years at Hewlett-Packard Laboratories in Palo Alto, California. His recent work has been concerned with component-base software engineering of distributed systems, particularly Web-based systems. He is a coauthor of Programming Language Concepts (John Wiley, 1998), Fundamentals of Software Engineering (Prentice-Hall, 2002), and Software Architecture for Product Families (Addison-Wesley, 2000).

Addresses:
Distributed Systems Group
Technical University of Vienna
Argentinierstraße 8/184-1
A-1040 Vienna, Austria

Tel. +43(1) 58801-18410
Fax +43(1) 58801-18491
Web: http://www.infosys.tuwien.ac.at/Staff/mj

Tutorials

SPARK, an ``Intensive Overview''

Peter Amey & Janet Barnes, Praxis Critical Systems

SPARK is an annotated sub-language of Ada which is unambiguous and suitable for rigorous static analysis. The tutorial, which is extracted from the four-day ``Software Engineering with SPARK'' course will provide an intensive introduction to SPARK and the static analysis performed by the SPARK Examiner. Attendees will be encouraged to bring laptop computers on which the SPARK Examiner will be installed.

MaRTE OS: Bringing Embedded Systems and Real-Time POSIX Together

Michael Gonzalez Harbour & Mario Aldea, University of Cantabria

MaRTE OS is a free software implementation of the POSIX minimum real-time system profile. It is designed for embedded systems and provides a development environment for Ada, C, or mixed language real-time applications. The tutorial will describe the main features of MaRTE OS, its architecture and performance, and the details on its development environment.

In addition, the tutorial will discuss the main real-time operating system services defined in the POSIX.13 minimum real-time profile. These services allow application developers to write portable applications that meet their real-time requirements, and that may be be implemented on small embedded systems.

Principles Of Physical Software Design in Ada95

Matthew Heaney

The tutorial addresses issues concerning the compilation of large software systems and presents many techniques for ameliorating the problems.

Most texts on software design concentrate almost exclusively on logical design, and provide only a cursory explanation of physical design. Discussions about types and objects are important, but there are also many pragmatic compilation issues that cannot be ignored. Unless care is taken, dependencies among modules often force a substantial recompile when seemingly innocuous changes are made. This can stymie development, especially for large systems that require hours (or even days) to rebuild.

Implementing Design Patterns in Ada95

Matthew Heaney

This tutorial addresses the question of what ``design patterns'' are and presents many advanced idioms for object-oriented programming in Ada95.

CORBA 3 and CORBA for Embedded Systems

S. Ron Oliver, Top Graph'X

The tutorial starts with an overview of CORBA 3 with emphasis on changes from CORBA 2. Thereafter it addresses CORBA principles, the Interface Definition Language (IDL), client programs, object (server) programs, CORBA Services, CORBA Facilities, and the CORBA Component Model (CCM). Several advanced features of CORBA 3, including Minimum CORBA and Real-Time CORBA, are also discussed. These topics are of particular interest when using CORBA in the area of embedded systems.

Using Open Source Hard- and Software to Build Reliable Systems

Joel Sherrill, OAR Corporation
Jiri Gaisler, Gaisler Research

A framework for the development of embedded systems based solely on open-source components is presented. The framework is based on the LEON SPARC-V8 processor, RTEMS real-time operating system, and the GNU Ada toolchain. The tutorial includes a discussion of the implications of applying the open source model to hardware and embedded systems software. An overview of the characteristics of real-time embedded systems, the cross development process, and the features of Ada95 that aid the development of real-time embedded systems is presented. A demonstration is made on how to configure the target processor, adapt the RTEMS operating system to custom boards, and develop Ada applications.

Cleanroom Software Engineering: An Overview

William Bail, MITRE & University of Maryland

Cleanroom Software Engineering is an approach to the development of software that emphasizes defect avoidance and that is strongly rooted in formal methods and mathematics. While not gaining the notoriety that other techniques have enjoyed, projects that have applied Cleanroom have experienced significant benefits, including low defect rates. It emphasizes multiple builds in an incremental model, with each build constructed using forms known as box structures. Verification of the structures is accomplished using correctness proofs, while software certification is based on usage models which facilitate statistical testing. Recent work has integrated Cleanroom with object-oriented models. In addition the SEI has released a Cleanroom Software Engineering Reference Model, providing an integrated set of work products and processes for organizations wishing to apply this technique.

Exceptions - What You Always Wanted to Know About Exceptions, But Were Afraid To Ask

Currie Colket, MITRE & ACM SIGAda

Exception processing has the power to detect serious problems in the execution of a program and return one back to a known safe state with high integrity. As such, it can be a very powerful tool for developing high quality software.

To be effective, exceptions and their handling must be addressed at the design level and not at the code level.This presentation will discuss several alternative approaches to addressing error handling in the design using exceptions.

Moreover the use of exceptions can be assessed via automated tools. Several analyses that can be performed on a program via automated tools so the program quality can be improved will be discussed. The tutorial will conclude by addressing proposed needs for exceptions resulting from the Exception Workshop held at Ada-Europe 2001.

Workshop: Standard Container Library for Ada

Both contemporary dominant general purpose programming languages, Java and C++, come equipped with a standard set of reusable containers, such as Maps and Sets. There are quite a few Ada libraries for these purposes, but there is little agreement on the exact details of a standard container library. There is however a general feeling, as can be witnessed on recent discussions on comp.lang.ada, that such a library is important for Ada's future. A standard container library is important for achieving many of Ada's goals, top among them the use of reusable components for efficient software engineering. Other important goals that can be served by a standard container library are educational uses and efficient implementation of common algorithms and data structures, which is important for real-time systems. Designing a useful standard container library for Ada is a difficult task, as the language is used in a wide variety of different domains, with different and at times conflicting demands. Hence the need for debating and elaborating the issues among a group of interested Ada users.

Scope The workshop will be confined to container library issues, and will not address more general questions regarding the Ada standard library. The library designed is intended to be a collection of abstract data types, data structures (i.e., concrete data types), and common algorithms, useful for sequential programming. Possible candidates for inclusion in the library will be judged according to this mission statement. Concurrent versions of the containers will be considered to the extent that their inclusion will not interfere with the mission statement, either by complicating the design, or by imposing unacceptable run-time overhead. Part of the design process will be dedicated to exploring this issue, including exploring how this can be related to the extendibility of the library. The suggested design should be based on standard Ada95 features.

Aims of the workshop

The workshop participants should attempt to come up with two deliverables:

A set of core requirements that are deemed by the majority to be critical for applicability of the standard container library to their needs.
A high level design for the library.

The workshop should help in forming a working group dedicated to implementing a reference implementation of the proposed library.

The results of the workshop should form the basis for a recommendation which would lead to the adoption of a standard container library as part of the Ada standard library, in the next revision of the Ada language. A first step towards this goal would be to achieve compiler vendor support.

Participation Participation to the workshop is limited to 25--35 individuals and is by invitation upon acceptance of a submission. Participants should submit brief position papers (1 or 2 pages) to Ehud Lamm (ehudla@oumail.openu.ac.il), including link to relevant source code if at all possible. Experience reports, full research papers, etc. can be referenced and circulated separately where appropriate. All papers will be made available to workshop participants before the workshop. The workshop will include talks based on the submitted papers and intensive shepherded design sessions. The submissions and a workshop summary will be published in the Ada User Journal.

Participants are encouraged to present a list of requirements from the standard container library, categorized into essential, important, and nice to have. Part of the workshop session will be dedicated to establishing a master list of requirements.

Suggested programming problems for evaluating data structure libraries

Submissions

Position Papers deadline: 1 June, 2002

Workshop co-chairs

John English (U. of Brighton)
Ehud Lamm (Open U. of Israel)

Date

The workshop is to be held on Friday, June 21st, 2002, in the morning.

Last Changed: Fri Jun 14 22:22:18 CEST 2002 from Document Version 1.8
Contents of this page: Johann Blieberger
Webmaster: Heinz Deinhart