BeyeNETWORK - The Global Vision for BI and Beyond

A critical crossroads in successful system delivery is the intersection between the systems development methodology and the enterprise architecture framework. It is at this intersection that the two must be aligned to achieve the combined benefits of both – improved budgeting accuracy, effective planning, and ability to move the enterprise to its committed future state.

Let's start by looking at the definitive enterprise architecture framework, the Zachman framework at Zifa.com. It is highly adaptive, as has been proven out over the last 20 years. It is methodology neutral. There are no sequence numbers in the cells. Each cell, or prime, in the framework provides a building block that is perpetual and reusable.

 
Figure 1: Zachman Framework

An example system development life cycle (SDLC) is shown in Figure 2. The key activities moving from left to right are initiation, analysis, design, construction, acceptance, and implementation.

 
Figure 2: Example SDLC

In order to align the systems development methodology and the enterprise architecture framework, let's take a sideways view of Zachman framework and use the SDLC shown in Figure 2 as an example. The result is shown in Figure 3.

Combinations of the primes for projects provide temporal composites that play a key role, when moving through the SDLC. The overall progression through the framework for the SDLC is from the contextual to the conceptual, then logical, and on to physical as described in the following paragraph. Each stage represents a gateway. In our experience, the architecture review process greatly benefits from knowing what is expected at each of these gateways. The review flows more smoothly when it is criteria driven, calling out the framework artifacts.

Figure 3: SDLC and Framework Alignment

Starting at the far left in Figure 3, prior to entering initiation, the contextual view provides the scope as part the high-level business case or charter development. This scope is a composite of the first row primes from the framework for this project. The minimum composite from the first row is to link the strategies (column 6) to processes (column 2) and data (column 1). However, at this level, there should also be information available for the other three columns to complete the picture.

Moving on in the life cycle, but still prior to initiation, the conceptual view provides the proposed solution. Here the key composite is again focused on the strategies (column 6) to process (column 2) and data (column 1), but now further broken out to the conceptual level. The composite is designed in the context of the reusable enterprise primes, ensuring the project outcomes aggregate to enterprise solutions and the desired future state.
 
At this stage, the solution is independent of technology. Depending on how the funding model overlays the SDLC, it may be necessary to drop down briefly to the logical and physical views to assess the impacts to existing systems and specify the required skill sets.

The analysis stage drives out the logical system model. The key pieces from the framework include a logical data model, data flow diagram, component diagram, and high-level sequence diagram. The logical data model is a subset of the prime focused on the portion that pertains to the scope of the project. It is fully attributed and at a level to provide all the information needed as input to the next stage of the SDLC. The data flow diagram is a composite. It links the application functions (column 2) and the data entities (column 3) moving between them. The component and sequence diagrams are also composites. The first one combines application functions and the data entities, and the second one is a combination of function, data, and time (column 5).
 
Next, the physical model is created in the SDLC design stage and is the essential precursor to construction. The fundamental models from the framework are the physical data model, lower-level sequence diagram, class diagram, and deployment diagram. The physical data model is driven out from the logical model of the previous stage and factors in the physical constraints. The lower-level sequence diagram combines primes from time, function, and data. The class diagram spans function and data. The deployment diagram draws from network, function, and data.

At each step of the system development life cycle, it is necessary to engage correctly with the enterprise architecture framework as described previously. It is at the critical crossroads between the two that essential artifacts are aligned to the development process. Moving through the life cycle, the architectural artifacts from the framework evolve from the contextual to the conceptual, then logical and down to the physical level prior to construction.

It is in this way that the accuracy of the budget and plans can be improved at each stage. The additional precision and detail aids in refining the duration and costs estimates. At the same time, since each level of detail specifies how the target state is achieved, you can ensure the resulting implemented solution moves the company forward toward the desired future state. The intersection of the systems methodology and enterprise architecture framework is truly a critical crossroads in driving clarity and refinement in successful system delivery.

If you found this article helpful and would like to receive the latest insights each week from the experts featured on the BeyeNETWORK, please subscribe to the Bill Inmon Newsletter.