More talks in the program:
11:30 - 12:15
Almost all modern tech stack components play along very well with cloud native workflows. Databases are an exception. They often get special treatment excluding them from normal DevOps operations and patterns like branching, pull requests, and high velocity iteration.
In this session, we will talk about some of the specific challenges of databases that have led to this special treatment. With schema changes in MySQL as a primary use case, we will talk about methods that have been used to overcome some of these challenges. We’ll go into detail about the schema change migration flow at GitHub, some of the problems encountered there, and how we built the tool gh-ost solve it.
We’ll conclude by talking about the next generation of schema change technology powered by Vitess, and discuss how PlanetScale leverages it to create a flow that treats database deployments like code deployments, including the ability to roll back a bad change.