🔗 We replaced Redis with MySQL for inventory reservations—and it scaled

Shopify faced challenges managing inventory reservations under peak conditions, especially with record sales of $5.1 million per minute on Black Friday 2025. Their previous setup, using Redis for oversell protection, struggled with multi-location awareness and atomic operations. By transitioning to MySQL and utilizing the SKIP LOCKED feature, they restructured their reservations to one row per inventory unit.

This change enabled them to maintain ACID properties across transactions, significantly improving scalability without the pitfalls of overselling or underselling. Key lessons included optimizing primary keys to minimize lock contention and enhancing connection visibility to better manage database load. Ultimately, by running both systems in shadow mode while transitioning, Shopify ensured correctness and performance before fully switching to MySQL, allowing for a smoother integration without impacting customers.