Shopify's engineering team says it swapped Redis for MySQL to handle inventory reservations — the machinery that decides whether the last pair of sneakers in a warehouse is really still yours when you hit checkout — and that the new setup scaled.
The claim comes from a post on Shopify's engineering blog titled Shopify replaced Redis with MySQL for inventory reservations–and it scaled, which reached the front page of Hacker News, where it drew 146 points and 81 comments.
The direction of the change is what makes it notable. Redis is an in-memory data store prized for raw speed, and it is a standard tool for absorbing bursts of traffic faster than a traditional database can. MySQL is a decades-old relational database, usually cast as the slower but sturdier option. Prevailing instinct in large-scale web engineering runs the other way: pull the hottest, most contested operations out of the database and push them into a fast cache. Shopify, by the title's account, went backward and came out ahead.
Beyond that headline claim, the source material here does not spell out the numbers, the architecture, or the trade-offs behind the migration. Those details live in Shopify's own write-up, and the volume of discussion on Hacker News suggests engineers are actively arguing over them rather than nodding along.
Why it matters: the unglamorous database decisions behind a checkout button determine whether carts break on the busiest shopping days of the year, and when a company operating at Shopify's scale publicly reverses a widely held architectural assumption, other engineering teams have reason to re-examine defaults they adopted without ever testing.