Automated Database Deployments – Principles for Success

Brian Hibberd

When it comes to automated database deployments, I’ve been on quite the journey, from home-grown scripting to SQL Server Data Tools, then Redgate SQL Change Automation and now Flyway. Through it all, a handful of principles have emerged – the “non-negotiables”, the secrets to our success. And I want to share them with you.

We’ll touch on the high-level, like the difference between declarative and imperative deployment paradigms, and the reason declarative tooling is tricky against large datasets. We’ll look at how most automated deployment frameworks apply changes to their targets. And we’ll get into the nitty-gritty, like changing large indexes or handling bugs in your deployment scripts.

Whether you’re deploying to SQL Server or PostgreSQL, whether you’re deploying through Octopus or Azure Devops, these principles are likely to be helpful.