The Postgres Mistakes that we all still make

Abstract: Let's take a look at all of the things we do in Postgres (even after 30 years) that cause repeat trouble. What workloads do we run that cause excessive bloat and how can we avoid it? When connection storms take over the database, how do we protect against them? How does idle in transaction keep you from getting work done, and how can you move past it? We'll take a look at the common lost battles and build a plan so that you can avoid these traps before you hit them. Speaker/ Speakers Bio: Scott Mead is a Senior Staff Engineer at Databricks, where he works on managed PostgreSQL at scale. He has 22+ years of experience with PostgreSQL as a user, DBA, developer, and contributor, and throughout his career, his focus has stayed on production performance — the systems, patterns, and pitfalls that show up when Postgres runs at scale in the real world.