ConcernBASE

Motivation
Since the number of organizations using software infrastructures to run their core business grows, society is becoming increasingly dependent on long-running and large software-intensive systems. Complex software-intensive systems are often constructed by putting different parts together rather than developed from scratch. Therefore, we need a means to provide blueprints for constructing such systems, to help stakeholders understand, manage and analyze fundamental system properties, as well as to provide mechanisms that allow software engineers to communicate and reason about those properties. This is achieved by what we refer to as software architecture.

On one hand, representing software architectures in an unambiguous and explicit way has been characterized as a critical issue in the design and construction of any complex software system. On the other hand, current software architecture concepts and mechanisms, including architecture description languages (ADLs) and architecture-based methods, do not provide means for simultaneously separating multiple kinds of concerns at different levels of abstraction.

We believe that a software architect, to be able to provide an architecture description that reflects all architecturally significant aspects of a system, requires appropriate mechanisms that allow us to separate various kinds of architectural concerns which possibly overlap one another. In this project, we take a new direction that puts emphasis on exploiting multi-dimensional separation of concerns in software architecture representations while focussing on large-scale software construction, software component development and the combination of both.

Objectives
The ConcernBASE project aims at providing a software engineering approach that complements the abstractions and mechanisms found in current ADLs, allows for simultaneous separation of overlapping concerns in software architecture descriptions and provides flexible means of supporting architecture-centered software development.

In this project, we consider the following main fields:

  • Conceptual framework issues: characteristics of standards, abstractions, notation and tools that influence composition and decomposition mechanisms and their combination in software architecture.
  • Methodological issues: characteristics of the systematic integration of software architecture descriptions with other software development artifacts.
Members
Shane Sendall
Related Areas
Multi-Dimensional Separation of Concerns and Hyperspaces
Software Architecture

Contributions
Mohamed Mancona Kandé and Alfred Strohmeier, "Towards a UML Profile for Software Architecture,'' UML'2000 - The Unified Modeling Language: Advancing the Standard, Third International Conference, York, UK, October 2-6, 2000, Stuart Kent and Andy Evans Ed.), LNCS Lecture Notes in Computer Science), no. 1939, 2000, pp. 513-527.

Mohamed Mancona Kandé and Alfred Strohmeier, "On The Role of Multi-Dimensional Separation of Concerns in Software Architecture''. OOPSLA’2000 Workshop on Advanced Separation of Concerns in Object-Oriented Systems, Minneapolis, Minnesota USA, October 2000.

EPFL | IC | LGL | Research
URL: http://lgl.epfl.ch/research/concernbase/index.html
Last modified 9/11/2001, M. Kandé.