This session will provide an overview of the 3 phased approach to database reengineering as described in Chapter 9 of the Modernizing IBM i Applications Redbook. This begins with the technique for transparently re‐engineering existing databases using the “measure twice, cut once approach”. In other words, exploiting Data Centric constructs with little or no impact to existing programs. The session will cover the process of building a solid database foundation using a physical, virtual and data access layer framework. The use of IBM development tools (i.e. Infosphere Data Architect, IBM Data Studio and Rational Developer) will be discussed. In addition, minimizing the impact of change via the use of bridge techniques such as “surrogate” files (DDS and SQL), RPG Open Access, Global Variables with Triggers, and Instead of Triggers (i.e. creating updateable SQL Join views) will be discussed.
The following is the agenda for this session:
• Overview of the IBM Reengineering Strategy
• Reverse Engineering an existing Database
• Refactoring Data Access
• Restructuring the Tables
Over the last 4 years or so, the IBM Lab Services DB2 for i team has been heavily involved in contributing to, and/or completely writing, several new IBM Redbooks including “Modernizing IBM i Applications from the Database up to the User Interface and Everything in Between”, “Row and Column Access Control (RCAC) Support in IBM DB2 for i” and most recently, “SQL Procedures, Triggers, and Functions on DB2 for i” (aka SQL Routines). The Friday sessions, and 3 of the Saturday sessions, are based on these Redbooks.