Model-Driven Enterprise and Software Architecture

Elements of a solution

There are many challenges in Enterprise Architecture, each potentially addressed by  slightly different tools to best resolve a particular kind of issue. However, when combined into a consistent overall approach, these tools bring a new coherence and confidence to a difficult undertaking, helping bridge the gap between business and IT. A consistent approach should ask the right questions, uncover key issues and insights, and relate disparate viewpoints into an understandable whole.

Our approach is called MAp. It is built on and extends our previous work on component modeling with Catalysis, integrating key aspects of the OMG's MDA, UML, and EDOC, the Reference Model for Open Distributed Processing (RM-ODP), SEI (Software Engineering Institute) work on software architecture, Problem Frames, and Goal Modeling. It covers the most important issues in Software and Enterprise Architecture.

  • Enterprise viewpoints: Viewpoints for the enterprise must cover business goals and processes, portfolio management, cross-business perspectives, and migration planning, in additional to application-centric viewpoints on business and technical components. Key questions are asked, and issues and insights uncovered, early.
  • Business goals: An effective enterprise architecture must span the business view, logical application architecture, platform-specific realizations, and deployed solutions. It's ultimate success is defined by the business value of the systems it delivers i.e. how well they meet business goals. Hence goals must be described unambiguously.
  • Fractal components and services: Applications, federations of applications, components, and services can all be viewed as points in a spectrum of fractal architectural elements. Shed a few buzzword pounds and gain in simplicity and consistency of approach.
  • Architecture-centric process: The processes used to define, develop, and evolve systems should be primarily aligned with the architectural structures of those systems, rather than be dominated by an orthogonal "phase" structure.
  • Architecture vocabulary: A clearly and consistently defined vocabulary with which to describe, analyze and draw inferences, and evaluate and choose architectures, is a significant benefit to most large-scale architectural efforts.
  • Portfolio management: Many important decisions about priority, direction, and architectural strategy should cross lines of business, applications, and projects and programs.
  • Information content: Every enterprise prizes its business information. Key elements of that information, starting from potential contents of a management dashboard and key-performance indicators, through detailed information exchanges between applications and components, can all be uncovered in a straightforward and consistent way.
  • Model-driven process: Models, used effectively, can add a lot of value to activities from defining goals to the integration of systems. But diagrams are not an end in themselves, and by shaking off diagram-religion in favor of the underlying shared conceptual language, you can start to define and benefit from models very early.
  • Variable zoom and focus: It is essential to be able to zoom out to get a wider birds-eye view of some domain or system, or zoom in to a narrower scope to be worked in more depth. Similarly, it is important to employ variable precision so even zoomed out views can retain as sharp a focus as necessary.
  • Enterprise lifecycle: In the context of enterprise architecture, the lifecycle of an application includes concept, implementation, deployment, production, evolution, and retirement. Analysis of as-is and development of to-be architectures frame one part of this lifecycle. Migration planning, and an architectural language centered on "change", is equally important.
  • Consistent architectural styles: An architectural style defines a set of acceptable or preferred architectures. By consistently using architectural styles within and across viewpoints, from business goals and domain models through technical architecture and deployments, you can leverage best practice from industry as well as within your organization.

Learn more about how we can help weave these together to bring renewed coherence and confidence to your efforts.