Avoid Toll Roads: Deploying Declarative Database Schema Changes
summary
Atlas is an open source tool that manages database schema declaratively, connecting to live databases, inspecting schema with a schema loader, calculating diffs and topologically ordered execution plans, and turning them into SQL. Demos show the CLI creating migrations, a lint step that flags destructive changes, a precomputed automatic migration planning mode that emits reviewed SQL, and a Kubernetes operator that enforces diff policies like forbidding drops and creating indexes concurrently.
AI-written summary · about
chapters (16)
- 0:00Introduction and speaker background
- 1:05Agenda and talk overview
- 1:48Database schema basics and challenges
- 3:58Gap in schema management tools
- 7:33Imperative vs declarative models
- 9:15Navigation metaphor for schema changes
- 11:28Live demo of declarative schema changes
- 15:45Risks of declarative workflows
- 16:48Avoiding toll roads in schema changes
- 17:54Automatic migration planning
- 19:21Demo of migration diff and linting
- 21:49Diff policies for safer changes
- 23:13Demo of diff policies
- 24:22Kubernetes operator for schemas
- 28:21Future of database automation
- 30:08Q&A and closing remarks
related talks
31mBeyond Alembic and Django Migrations
25mUpdating Databases the GitOps Way
31mHelp! My Database Is The Bottleneck
27mStrategies for Reliable Database Schema Migrations
05mMaking Database Changes a Tier one Artifact
08mDatabase as Code. Not only migrations
28mSchema-as-Code: Developer platforms and database schema changes
57mAlgebraic Databases