The Life-Cycle Approach Astir
As IT professional you are all too familiar with the traditional system development life cycle (SDIC) know how to begin with a project plan, move into the requirements analysis phase, then into the design, construction, and testing phases, and finally into the implementation phase.
The life cycle approach accomplishes all the major objectives in the system development process. it enforces orderliness -and enables a systematic approach to building computer systems. The life cycle methodology breaks down the project complexity and removes any ambiguity with regard to the responsibilities of project team members. It implies a predictable set of tasks and deliverables. That the life cycle approach breaks down the project complexity is alone reason enough for this approach to be applied to a data warehouse project. A data warehouse project is complex in terms of tasks, technologies, and team member roles. But a one-size-fits-all life cycle approach will not work for a data warehouse project. Adapt the life cycle approach to the special needs of your data warehouse project. Note that a life cycle for data warehouse development is not a waterfall method in which one phase ends and cascades into the next one.
The approach for a data warehouse project has to include iterative tasks going through cycles of refinement. For example, if one of your tasks in the project is identification of data sources, you might begin by reviewing all the source systems and listing all the source data structures. The next iteration of the task is meant to review, the data elements with the users. You move on to the next iteration of reviewing the data elements with the database administrator and some other IT start. The next iteration of walking through the data elements one more time completes the refinements and the task. This type of iterative process is required for each task because of the complexity and broad scope of the project. Remember that the broad functional components of a data warehouse are data acquisition, data storage, and information delivery. Make sure the phases of your development, life cycle wrap around these functional components. Figure 4-3 shows how to relate the functional components to SDLC.
Figure 4-3 DW functional components and SDLC
As in any system development life cycle, the data warehouse project begins with the preparation of a project plan. The project plan describes the project, identifies the specific objectives, mentions the crucial success factors, lists the assumptions, and highlights the critical issues. The plan includes the project schedule, lists the tasks and assignments, and provides for monitoring progress. Figure 4-4 provides a sample outline of a data ware-house project plan.