I love Redis. It's one of those technologies that is so obvious it makes you wonder why it took so long for someone to build it. Predictable, performant, and adapatable, it's something I've come to use more and more over the last few years. It's also no secret that Sentry is run primarily on PostgreSQL (though it now also relies on a number of other technologies).

A little over a week ago I gave a keynote at Python Nordeste. At some point it was suggested I give a lightning talk. I decided instead I'd talk about some of the cool hacks we use to scale Sentry, specifically with