Thursday, February 22 • 3:20pm - 3:50pm
DevOps Comes to the Database

Sign up or log in to save this to your schedule and see who's attending!

Feedback form is now closed.
Databases are special because they’re stateful, and DevOps reflects this: in practice, databases are a lot harder to “DevOps” (as a verb) than other parts of the stack. It’s easy to “DevOps the infra” and “DevOps the web frontend” and so on, because you can just use a declarative tool to make it be the way it should be, and if one of them goes belly up you just kill and replace it. But you can’t do that with the database, and what often happens is orgs get really open-minded about EVERYTHING BUT THE FREAKING DATABASE.
And who suffers? One person often becomes the special database gatekeeper, like Gandalf saying “you shall not pass!” and they suffer. But you know who else suffers? Every developer who now has to coordinate with, get review from, and get permission from the database gatekeeper. Their productivity also suffers.
There’s a better way. It requires cultural support, it requires tooling (like most of DevOps, although it’s not about tools you can’t actually DO it without tools), and it requires a bit of courage. You CAN build tools and processes that help you move faster and be more agile even when – especially when – very large, heavily loaded, mission critical databases are involved. This talk is about how I’ve done it and how my customers have done it.


Thursday February 22, 2018 3:20pm - 3:50pm