Meta-Modeling

Models in system design follow the general tendency in electronics in terms of size, complexity and difficulty of maintenance. While a model should be a manageable representation of a system, this increasing complexity sometimes forces current CAD-tool designers and model writers to apply modeling techniques to the model itself. Model writers are interested in instrumenting their model, so as to extract critical information before the model is complete. CAD tools designers use internal representations of the design at various stages. The complexity has also led CAD-tool developers to develop formal tools, theories and methods to improve relevance, completeness and consistency of those internal representations. Information modeling involves the representation of objects, their properties and relationships.

Performance Modeling

When it comes to design choices and trade-offs, performance is generally the final key. However performance estimations have to be extracted at a very early stage in the system design. Performance modeling concerns the set of tools and techniques that allow or help the designer to capture metrics relating to future architectures. Performance modeling encompasses the whole system, including software modeling. It has a strong impact on all levels of design choices, from hardware/software partitioning to the final layout.

Information Modeling

Specification and formalism have in the past traditionally played little part in the design and development of EDA systems, their support environments, languages and processes. Instead, EDA system developers and EDA system users have seemed to be content to operate within environments that are often extremely complex and may be poorly tested and understood. This situation has now begun to change with the increasing use of techniques drawn from the domains of formal specification and database design. This section of this volume addresses aspects of the techniques being used. In particular, it considers a specific formalism, called information modeling, which has gained increasing acceptance recently and is now a key part of many of the proposals in the EDA Standards Roadmap, which promises to be of significance to the EDA industry. In addition, the section looks at an example of a design system from the point of view of its underlying understanding of the design process rather than through a consideration of particular CAD algorithms.

Meta-Modeling: Performance and Information Modeling contains papers describing the very latest techniques used in meta-modeling. It will be a valuable text for researchers, practitioners and students involved in Electronic Design Automation.