The key question
Moving to AWS can change the operating model without removing Oracle. Moving to PostgreSQL also changes database capabilities and often the application. The two initiatives therefore have different evidence and acceptance criteria.
What to inventory
- Clarify the business objective: operating model, modernisation or engine change.
- Capture interfaces, outage windows and data-transfer requirements.
- Mark application components that cannot change and key technical dependencies.
Design the target deliberately
Assess whether a direct engine change or a staged approach is safer. An intermediate step can decouple workstreams, but it also creates extra migration and operating effort. Document which specific risk the intermediate state is meant to reduce.
A concrete validation step
For each stage, record objective, benefit, added effort, remaining dependencies and exit criterion. That prevents a “temporary” architecture from becoming permanent by accident.
Evidence required before sign-off
- Give every stage its own acceptance target.
- Define reconciliation and rollback conditions before cutover.
- Test network paths and external systems under realistic conditions.
- Make post-move responsibilities explicit.
Primary sources & further reading
These recommendations are engineering guidance. Specific options depend on source and target versions, privileges and operating model.