Case Study: Slack scales MySQL with Vitess from Cloud Native Computing Foundation

How Slack leverages Vitess to keep pace with ever-growing storage needs

Slack, a collaboration platform with millions of users, faced significant database scaling challenges with its legacy MySQL infrastructure. The company's busiest database hosts were overwhelmed by the load from large customers, while thousands of other hosts remained idle, creating an inefficient and operationally taxing system. This prompted Slack to seek a new solution with the Cloud Native Computing Foundation's Vitess.

The solution implemented by the Cloud Native Computing Foundation involved migrating to Vitess, a MySQL scaling and management platform. This allowed Slack to move away from its rigid, workspace-based sharding to a more flexible "shard by anything" model. The results were highly successful: 45% of Slack's query load, amounting to 20 billion queries per day, now runs on Vitess with only a negligible 1ms of added latency. The platform now sustainably powers critical features like @-mentions and reactions, and Slack plans to migrate all MySQL databases to Vitess.

Slack

Michael Demmer

Principal Engineer