Automating The Design Of Computer Systems

Addressing the issues of engineering design in computer architecture, this book describes the design and implementation of MICON, a system for automating the synthesis of small computers.

This introductory book discusses how to plan and build useful, reliable, maintainable and cost efficient computer systems for automated engineering design. The book takes a user perspective and seeks to bridge the gap between texts on principles of computer science and the user manuals for commercial design automation software. The approach taken is top-down, following the path from definition of the design task and clarification of the relevant design knowledge to the development of an operational system well adapted for its purpose. This introductory text for the practicing engineer working in industry covers most vital aspects of planning such a system. Experiences from applications of automated design systems in practice are reviewed based on a large number of real, industrial cases. The principles behind the most popular methods in design automation are presented with sufficient rigour to give the user confidence in applying them on real industrial problems. This book is also suited for a half semester course at graduate level and has been complemented by suggestions for student assignments grown out of the lecture notes of two postgraduate courses given annually or biannually during the last ten years at the Product development program at the School of Engineering at Jönköping University.

The contributions in Automating Software Design provide substantial evidence that AI technology can meet the requirements of the large potential market that will exist for knowledge-based software engineering at the turn of the century. They are divided into sections covering knowledge-­based tools for large software systems, knowledge-based specification acquisition, domain-oriented program synthesis, knowledge compilation, knowledge-based program optimization, formal derivation systems, and cognitive and planning approaches to software design. Partial Contents: Knowledge-Based Software Engineering: How and Why Did We Get Here? The Evolution of Very Large Information Systems. LaSSIE: A knowledge-Based Software Information System. Reducing the Complexity of Formal Specification Acquisition. Software Reuse and Refinement in the IDeA and ROSE Systems. Data Relationships and Software Design. Scientific Programming by Automated Synthesis. Synthesizing VLSI Routing Software from Specification. A Divide-and-Conquer Approach to Knowledge Compilation (the KBSDE project). Program Improvement by Automatic Redistribution of Intermediate Results: An Overview. Concurrent Software Production. Design Principles for an Interactive Program Derivation System. The Structure and Design of Local Search Algorithms. Automating Algorithm Design Within a General Architecture for Intelligence. Software Engineering in the Twenty-First Century.

Design automation of electronic and hybrid systems is a steadily growing field of interest and a permanent challenge for researchers in Electronics, Computer Engineering and Computer Science. System Design Automation presents some recent results in design automation of different types of electronic and mechatronic systems. It deals with various topics of design automation, ranging from high level digital system synthesis, through analogue and heterogeneous system analysis and design, up to system modeling and simulation. Design automation is treated from the aspects of its theoretical fundamentals, its basic approach and its methods and tools. Several application cases are presented in detail. The book consists of three chapters: High-Level System Synthesis (Digital Hardware/Software Systems). Here embedded systems, distributed systems and processor arrays as well as hardware-software codesign are treated. Also three special application cases are discussed in detail; Analog and Heterogeneous System Design (System Approach and Methodology). This chapter copes with the analysis and design of hybrid systems comprised of analog and digital, electronic and mechanical components; System Simulation and Evaluation (Methods and Tools). In this chapter object-oriented Modelling, analog system simulation including fault-simulation, parameter optimization and system validation are regarded. The contents of the book are based on material presented at the Workshop System Design Automation (SDA 2000) organised by the Sonderforschungsbereich 358 of the Deutsche Forschungsgemeinschaft at TU Dresden.

1 INTRODUCTION These proceedings are the result of a conference on Automating Systems Development held at Leicester Polytechnic, England on 14 to 16 April 1987. The conference was attended by over 170 delegates from industry and academia and it represents a comprehensive review of the state of the art of the use of the computer based tools for the analysis, design and construction of Information Systems (IS). Two parallel streams ran throughout the conference. The academic, or research, papers were the fruit of British, European and Canadian research, with some of the papers reflecting UK Government funded Alvey or European ESPRIT research projects. Two important touchstones guided the selection of academic papers. Firstly, they should be primarily concerned with system, rather than program, development. Secondly, they should be easily accessible to delegates and readers. We felt that formal mathematical papers had plenty of other opportunities for airing and publication. The second stream was the applied programme; a set of formal presentations given by leading software vendors and consultancies. It is clear that many advances in systems development are actually applied, rather than re search led. Thus it was important for delegates to hear how leading edge companies view the State of the Art. This was supported by a small exhibi tion area where certain vendors demonstrated the software they had intro duced in the formal presentation.

Neurocomputing for Design Automation provides innovative design theories and computational models with two broad objectives: automation and optimization. This singular book: Presents an introduction to the automation and optimization of engineering design of complex engineering systems using neural network computing Outlines new computational models and paradigms for automating the complex process of design for unique engineering systems, such as steel highrise building structures Applies design theories and models to the solution of structural design problems Integrates three computing paradigms: mathematical optimization, neural network computing, and parallel processing The applications described are general enough to be applied directly or by extension to other engineering design problems, such as aerospace or mechanical design. Also, the computational models are shown to be stable and robust - particularly suitable for design automation of large systems, such as a 144-story steel super-highrise building structure with more than 20,000 members. The book provides an exceptional framework for the automation and optimization of engineering design, focusing on a new computing paradigm - neural networks computing. It presents the automation of complex systems at a new and higher level never achieved before.

Computer systems based on the notion of the computer as assistant have recently become the focus of intense interest. The expanding role of the computer in everyday life and the growing number of untrained users make it necessary to think about new ways of dividing labor between humans and machines. Future systems must take on more tasks and perform them more competently and autonomously than existing systems. If they are to be adequately flexible and responsive to complexity, they cannot automate their performance completely. The aim of designers should be to create computer systems with capabilities similar to those of good assistants in the real world. Effective assistance has many characteristics. An assistant is expected to be competent in some domains of expertise, to know the limits of his/her knowledge, to be able to process inexact instructions from clients, to adjust to and learn from them, to explain his/her behavior and suggestions, and to support clients in communication and cooperation with other people. This book believes that such capabilities can be built into computer systems. To that end, the chapter contributors discuss the concepts and methods--particularly from the fields of artificial intelligence and computer-supported cooperative work (CSCW)--that they have drawn from to develop successful system prototypes. They present several of these prototypes including assistants for graphics design, knowledge discovery in data bases, coordination support, organizational memory, user interface design, and knowledge base construction. As such, this volume helps map out the future for all those involved in computer systems design.