Getting Relational Database from Legacy Data-MDRE Approach

Abstract

The previous management information systems turning on traditional mainframe environment are often written in COBOL and store their data in files; they are usually large and complex and known as legacy systems. These legacy systems need to be maintained and evolved due to several causes, including correction of anomalies, requirements change, management rules change, new reorganization, etc. But, the maintenance of legacy systems becomes over years extremely complex and highly expensive, In this case, a new or an improved system must replace the previous one. However, replacing those systems completely from scratch is also very expensive and it represents a huge risk. Nevertheless, they should be evolved by profiting from the valuable knowledge embedded in them. This paper proposes a reverse engineering process based on Model Driven engineering that presents a solution to provide a normalized relational database which includes the integrity constraints extracted from legacy data. A CASE tool CETL: (COBOL Extract Transform Load) is developed to support the proposal.

Keywords: legacy data, reverse engineering, model driven engineering, COBOL metamodel, domain class diagram, relational database