COBOL to Java Migration - A UK Enterprise Guide 2026

Java is the most common destination for enterprise COBOL migration in the UK. It offers a mature ecosystem, strong typing, a vast library catalogue, and an extensive developer pool, making it ideal for organisations with critical COBOL on IBM mainframes. The Java Virtual Machine (JVM) provides portability, and the available tooling fits existing deployment, monitoring, and security infrastructure.

However, a successful migration requires adherence to a crucial rule: COBOL's packed-decimal (COMP-3) fields must map to Java's BigDecimal, not double or float. This ensures financial precision and avoids rounding errors in monetary calculations.

There are three main migration approaches:

1. Automated Conversion: Tools like Mecanik parse COBOL and generate Java code. While it can expedite the process, it requires manual refinement for embedded SQL, CICS interactions, and decimal precision considerations. This approach is suitable for large codebases where quick removal of COBOL dependency is prioritized.

2. Parallel Rewrite: In this approach, the Java system runs alongside the COBOL system, processing the same inputs and validating outputs until Java is deemed ready. The COBOL system is then decommissioned. This method is ideal for mission-critical systems like payments and payroll, where continuity cannot be compromised.

3. Incremental Migration (Strangler Fig): COBOL programs are replaced with Java equivalents one at a time, creating a hybrid system that eventually becomes purely Java. This approach works well for large, monolithic COBOL systems where a full rewrite is impractical. However, it demands careful interface design between the COBOL and Java components and can prolong the hybrid state.

For most UK enterprises, a hybrid approach combining automated conversion, parallel rewrite, and incremental migration is the most practical solution. The cost typically ranges from £200,000 to £800,000, and the migration process can take one to two years, with significant risks associated with the data access layer and undocumented business logic.