Configware (Reconfigurable Engineering)
Engineering Complex Computer Systems
Restructuring the Electric Power Industry: Emerging Issues, Methods and Tools
Scenario-Based System Development
Techniques for Secure Systems Development
Will Reconfigurable Computing overcome General Purpose Computing ?
Computing systems, like desktop, embedded, and others, need more and more add-on circuitry. In such a system meanwhile the major part of silicon area is special purpose: accelerators needed because the limitations of the microprocessor running the software. These add-ons bring most of the performance, so that the microprocessor is a tail wagging the dog.
Configware, the new Software Engineering Paradigm.
But now downloading accelerators, rather than designing them, has become feasible by new hardware platforms, being orders of magnitude more powerful than FPGAs. More and more blurring hardware/software boundaries stimulate advances in new directions of software technology. Configware is becoming a novel paradigm - as an alternative to software.
Compiler and language designers to the front.
Hardware has become soft. Dynamically reconfigurable hardware (configware) is evolving from tinkertoy approach to a novel programming paradigm, introducing a new class of instruction level parallelism (ILP). Novel architectures drive the need for new compilation techniques. Compiler and language designers must cooperate with hardware architects to exploit combinations of instruction level parallelism and concurrency by optimizing multi paradigm compilers for a novel class of inhomogeneous high performance systems including new and emerging add-on technology platforms.
Which business model?
But this is also a business model issue. Computing systems including a high performance general purpose dynamically reconfigurable accelerator could extend hardware product life time. Sure this is not a business model for a wintel type industry forcing people to throw away their hardware frequently. To which area of silicon-based industry would fit such a business model?
Minitrack Chairs:
| Reiner W. Hartenstein | Veljko Milutinovich |
| Universitaet Kaiserslautern | Department of Computer Engineering |
| Informatik (CS&E) | School of Electrical Engineering |
| Bau 12, Postfach 3049 | University of Belgrade |
| D-67653 Kaiserslautern, Germany | POB 35-54 |
| E-mail: hartenst@rhrk.uni-kl.de | 11120 Belgrade |
| Serbia, Yugoslavia | |
| Fax: +381-11-324-8681 | |
| Fax: +381-11-762-214 (at home) | |
| E-mail: VM@etf.bg.ac.yu |
Modern computer systems and applications embody many different characteristics and properties that are currently addressed, studied, and optimized independently. Nevertheless, although it is of basic importance to focus on these aspects independently, as a whole these properties feature a complex interrelationship, and thus a higher-level view of the complete project becomes mandatory.
While perhaps some of the earlier computer systems could be described, designed and implemented with a particular focus on one objective (such as fault-tolerance or timeliness), or using a single method (such as Structured Programming), it is very questionable whether such modern and future applications can be. Nowadays almost all electronic products are becoming more and more software based: complex computer systems are becoming common in many sectors, such as manufacturing, communications, defense, transportation, aerospace, hazardous environments, energy, health care, etc. These systems feature a number of different characteristics (such as distributed processing, heterogeneous computational paradigms, high speed networks, novel bus systems, or special-purpose hardware enhancements in general) and performance requirements (such as real-time behavior, fault tolerance, security, adaptability, development time and cost, long life concerns). The concurrent satisfaction of the systems requirements have a considerable impact on the hardware characteristics and vice-versa. The analysis of the complete project, as a whole, is a major point in the design of the computer system itself and plays a basic role throughout the entire system life.
The ECCS Minitrack will bring together industrial, academic, and government experts from these various disciplines, to determine how the disciplines' problems and solution techniques interact within the whole system. Researchers, practitioners, tool developers and users, and technology transition experts are all welcome.
Minitrack Chairs:
| Alberto Broggi | Mike Hinchey | Alexander D. Stoyen |
| Dipartimento di Ingegneria dell'Informazione | Real-Time Computing Labortory, CIS | President & CEO |
| Universita' di Parma | New Jersey Institute of Technology | 21st Century Systems, Inc. |
| I-43100 Parma | Newark, NJ 07102 | 420 Hardscrabble Road |
| ITALY | FAX: (201) 596-5777 | Chappaqua, NY 10514-3030 |
| +39-521-90-5707 | E-mail: michael.hinchey@ul.ie | Tel: (914) 769-2939 |
| FAX: +39-521-90-5723 | Fax: (914) 769-0949 | |
| E-mail: broggi@CE.UniPR.IT | E-mail: alex@21csi.com | |
| http://www.ce.unipr.it/~broggi |
The Pentium processor bug, the failure of the AT&T switching system, the deaths related to overdosing from the Therac 25 irradiation machine, and reported software failures in Airbus aircraft provide evidence of the risks taken when we use advanced computer systems to control critical systems. When these critical systems are operational, financial resources, corporate reputations, industrial secrets, human lives, and the environment are all at risk. The complexity of these systems prohibits the use of conventional design and analysis techniques to provide high levels of confidence in the correct operation of these systems. The application of formal design and analysis to these critical systems is essential to provide this high level of confidence, and thus reduce the risks involved in the use of these systems. It is this risk reduction that ties this minitrack to the proposed technical area.
The application of formal methods to the specification, design, verification and validation of advanced computer systems is part of the system engineering process in several trial projects throughout industry. The results of these trial projects are the focus of this minitrack. We will seek experts in formal methods technology who have had industrial application experience and the early innovators in industry who have adopted formal methods in practice. The reporting of the results and lessons learned will enable others to build on the successes and avoid the pitfalls and failures of others.
A representation of the topics that could be addressed by the minitrack:
Minitrack Chairs:
| Jim Alves-Foss | Ann E. Kelley Sobel |
| Computer Science Department | Systems Analysis Department |
| University of Idaho | Miami University |
| Moscow, Idaho 83844-1010 | Oxford, Ohio 45056 |
| phone: (208) 885-7232 | phone: (513) 529-7541 |
| dept. phone: (208) 885-6589 | dept. phone: (513) 529-5928 |
| fax: (208) 995-9052 | fax: (513) 529-1524 |
| E-mail: jimaf@cs.uidaho.edu | E-mail: sobelae@muohio.edu |
The past several years have seen marked changes in the institutional structures of the electric power industry in many countries. In most cases the driving forces for change have come from the governments. Often the objective is to introduce more competition and to increase private sector ownership of the electric supply industry. The overall aim regardless of the drivers, is to reduce the price of electricity and increase customer responsiveness by methods other than direct regulation.
Primary examples of institutional change include unbundling of services and equipment, the introduction of non-utility or third party generation, proposals for and implementation of new bidding or auction arrangements for pricing electricity and corporatisation, among others. In most cases each element includes ownership changes, a competitive market with new procedures and a re-vamping of the regulatory process. Each of these elements has and will continue to have its own impact on the planning, operation, cost and reliability of power systems. This minitrack is especially interested in identifying the effects and impacts of institutional change (proposed or in place). It is especially interested in new tools and methodology, that is, the technical underpinnings needed to transition the system from the old to the new. Therefore, papers on topics related to the affected elements of restructuring of an electric power industry will be entertained.
Minitrack Chair:
Scenario descriptions of user interactions are widely employed in the design and development of computer systems and applications. For example, scenarios are used in strategic management to concretize the complex uncertainties that inhere in planning for alternative futures, and to render envisioned futures more vivid and auditable. Scenarios are used in requirements engineering to uncover hidden requirements and tradeoffs, to verify and validate requirements, and to integrate analysis of functional and nonfunctional requirements. Scenarios are used in human-computer interaction to rapidly and inexpensively prototype and assess user tasks, to conduct analysis and design in a vocabulary that permits end-user participation, and to provide an orienting design representation couched in the actions and experiences of the people who will ultimately use the technology.
This mini-track builds upon the recent Dagstuhl Workshop on Scenario Management held in Wadern, Germany, 8-13 February 1998. Our goal is to bring together researchers in strategic management, requirements engineering, and human-computer interaction to further the establishment of broader and more effective engineering practices in scenario-based system development. For this mini-track, we seek papers that integrate scenario-based techniques in at least two of these areas.
Minitrack chair:
| John M. Carroll |
| Computer Science Dept |
| 660 McBryde Hall |
| Virginia Tech |
| Blacksburg, VA 24061 USA |
| Phone: 1-540-231-6931 |
| FAX: 1-540-231-6075 |
| E-mail: carroll@cs.vt.edu |
| http://www.cs.vt.edu/~carroll/ |
Software process improvement efforts rely on the successful integration of many technical, people, and methodological issues. This provides the opportunity for rich socio-technical and interdisciplinary research studies in addition to those studies that focus on process and/or enabling technology issues. Papers are welcome which deal with issues related to models, methodologies, metrics, organizational, management, people and team, and technical aspects of Software Process Improvement. Both conceptual and empirical work dealing with any stage of software process improvement initiatives is appropriate.
Areas of interest include:
1. models -- CMM, SPICE, TickIT, ISO9000-3, etc.
2. tools -- web, database, IT architectures, etc.
3. methodologies -- process definition, process modeling and representation, etc.; incremental development,
object-oriented development; etc
4. metrics -- goal-question-metric paradigm; mathematical modeling; complexity, size, cost, and quality metrics, etc.
5. organizational -- software development teams, organizational change, distributed and virtual development, contextual
analysis, ethnographic field research, organizational learning, etc.
6. technical issues -- risk analysis; software reuse; statistical quality control; education and training
Minitrack Chair:
| Gene McGuire |
| Associate Professor |
| Computer Science and Information Systems |
| Clark Hall, Rm. 123 |
| American University |
| 4400 Massachusetts Ave., N.W. 20016-8116 |
| Washington, D.C. 20016-8116 |
| phone: (202) 885-3268 |
| fax: (202) 885-1479 |
| E-mail: mcguire@american.edu |
This minitrack concerns software engineering techniques specifically addressing the development, verification, and validation of secure software systems.
Secure software systems are in the forefront again. Techniques originally developed in the 1970s and 1980s for specification and verification of multilevel secure operating systems and network devices are a foundation for today's multilevel networks.
The early security researchers did not envision the widespread public use of secure transaction systems for the Internet economy. Underlying the innocuous browser icons for secure transactions are authentication and privacy schemes that require systematic validation and verification. Security problems with browsers, as well as possible privacy intrusions such as cookies, are reported in the popular press. Internet-related magazines occasionally report on or publish security-related articles. Their editorial standards, however, can be quite low.
In particular, manuscripts will be solicited in the following areas:
Minitrack Chairs:
| Brent Auernheimer (primary contact) | Deborah Frincke |
| Computer Science M/S 109 | Computer Science |
| California State University | University of Idaho |
| Fresno CA 93740 USA | Moscow ID 83844-1010 USA |
| 209 278 2573 (voice) | 208 885 6589 (voice) |
| 209 278 4197 (fax) | E-mail:frincke@cs.uidaho.edu |
| E-mail: brent@CSUFresno.edu |
Please return to this page for pending updates (last updated 3/25/98)