Migrating 155 SQL Server Databases to PostgreSQL at Scale: A Health care Success Story

Large migrations fail when they’re treated as one big project instead of a repeatable process. In this talk, I’ll walk through a case study where we migrated 155 SQL Server databases to PostgreSQL in two months for a leading healthcare organization, covering about 200 TB of data, 100K+ tables, 10K+ stored procedures/functions, and 1M+ views and sub-objects. We’ll cover the assessment and compatibility decisions that made speed possible, the difficult scenarios we encountered, and the operational and application changes required for PostgreSQL to behave correctly under SQL Server workload expectations. You’ll leave with a proven checklist and a realistic view of what it takes to succeed at scale.

What we’ll cover:

• The non-negotiables for success at scale - assessment, scope control, automation, validation.
• Compatibility realities: where SQL Server behaviors differ and how to plan for it.
• how we categorized databases to sequence work and reduce risk.
• T-SQL to PostgreSQL conversion strategy - patterns, anti-patterns, and refactoring decisions
• Application impact patterns involving typical changes in queries, drivers, transactions, and error handling.
• Data migration strategy for large objects and minimizing downtime during cutover using native CDC agents of SQL Server.
• Correctness, performance, and operational acceptance criteria using relevant validations.
• Performance and plan regressions: what we tested, how we compared, and tuning workflow.
• Cutover strategy: low-downtime patterns, rollback planning, and stabilization checklist
• Where tools help, and the available Open-Source tools supporting SQL Server to PostgreSQL migrations.
• Production rollout learnings including cutover, rollback readiness, and post-migration stabilization.