Iteration Project Planning
The data warehouse should be implemented one subject area at a time. The order of the implementation cycles was determined during the business question assessment (BQA) stage. These were driven by the specific business questions to be answered in each iteration.
The business requirements discovered in BQA and the technical requirements defined in the architecture design stage should now be refined through additional interviews and focus sessions. These interviews should be focused on the specific subject area being implemented. Analysis should be completed to provide the detail necessary to design and implement this single population project.
The detailed work plan for the design and implementation stage should be developed. You should also estimate the cost based on the selected population project. The plan, along with costs, should be presented to the business users and management who are responsible for making the decision to move forward. Once you receive approval, you should be prepared to assemble the development team and begin work.
The high-level architecture design defines the overall strategy by which the data warehouse is developed. The detailed design maps out the implementation of those processes for the specific iteration. In this phase, the physical data warehouse model should be developed, the source data inventory should be updated to include all of the necessary information for this subject area implementation project and the meta data requirements should be defined.
Procedures must be designed to handle the following: warehouse capacity growth, data extraction/transformation/cleansing, data load/refresh, access, backup/recovery and testing. Upon completion of these design activities, a critical design review should be conducted with the customers and project owners before the project moves forward into the development phase.
The hardware, software and middleware components should be purchased and installed, the development and test environment should be established and the configuration management processes should be implemented. Programs should be developed to extract, cleanse, transform and load the source data and to periodically refresh the existing data in the warehouse. Each program should be individually unit tested against a test database with sample source data. Metrics should be captured for the load process, and the meta data repository should be loaded with transformational and business user meta data.
Canned production reports should be developed, sample ad hoc queries should be run against the test database and the validity of the output should be measured. Once the programs have been developed and unit tested, system functionality and user-acceptance testing should be conducted for the complete integrated data warehouse system. System support processes of database security, system backup and recovery, and data archiving should be implemented and tested as the system is prepared for deployment. The final step is to conduct a production readiness review prior to transitioning the data warehouse system into production. During this review, the system will be evaluated for acceptance by the customer organization.
Transition to Production
Now, you should prepare to move the data warehouse into the production environment. The production database should be created, and the extraction/cleanse/transformation routines should be run against the production operational system. The development team should work with the operations staff to perform the initial load of this data to the data warehouse and execute the first refresh cycle. The data warehouse programs and processes will then be moved into the production libraries and catalogs. Roll-out presentations and tool demonstrations should be given to the entire customer community, and end-user training should be scheduled and conducted. Also, the help desk should be established and put into operation.
Once the system has been moved into production, the new system should be positioned for ongoing maintenance through the establishment of a change management board and the implementation of change control procedures for future development cycles. As new iterations are being developed and deployed, it is critical to monitor the current system's performance, maintain contact with existing users and provide proper maintenance to the deployed system.
Using this approach, you can deliver tangible results within three to six months of project inception. The business users' needs are met because they have been involved throughout the entire process and they can now see the tangible results of your team's efforts. In the end, your developers and the business intelligence users should agree that using a proven methodology is the key to building a successful data warehouse.
Register or login for access to this item and much more
All Information Management content is archived after seven days.
Community members receive:
- All recent and archived articles
- Conference offers and updates
- A full menu of enewsletter options
- Web seminars, white papers, ebooks
Already have an account? Log In
Don't have an account? Register for Free Unlimited Access