A Version Control Strategy for Branch-based Database Development


Our existing database development and release process was holding back team productivity and causing frequent deployment delays. It was manual and error prone. Developers worked on a shared development database, with all changes committed to a ‘monolithic’, centralized version control repository. The dependencies this introduced, and likelihood of conflicting changes, meant database development was restricted to a ‘serial’ process, tackling their backlog one task at a time.

We set out on a « mission » to bring support for parallel development workstreams and process automation to the database. This article explains how we achieved the first milestones:

Migrating our centralized, ‘monolithic’