How YouTube Was Able to Support 2.49 Billion Users With MySQL

Vitess MySQL:
• They use Vitess as an abstraction layer for MySQL
• They run a sidecar server to manage MySQL instances
• They rewrite expensive SQL queries by adding the LIMIT clause
• They cache popular data on the sidecar server to prevent the thundering herd
• They use a proxy server to route queries to the right MySQL shard
• They do connection pooling using a proxy server
• They use a distributed key-value database to store shard metadata
• They use an HTTP server to update the key-value database
• They wrote Vitess in Go and open-sourced it
----------------------------------------------------------------
SUBSCRIBE OR I STOP WRITING
╔═╦╗╔╦╗╔═╦═╦╦╦╦╗╔═╗
║╚╣║║║╚╣╚╣╔╣╔╣║╚╣═╣
╠╗║╚╝║║╠╗║╚╣║║║║║═╣
╚═╩══╩═╩═╩═╩╝╚╩═╩═╝
→ ‪@systemdesignone‬
--------------------------------------------------------------------