Postgres Summit US2026

PostgreSQL is famously stable. Applications depend on that stability: SQL behavior, extensions, replication, authentication, planner behavior, and operational expectations are expected to remain predictable over time.

But in production, the ground beneath Postgres is constantly moving. The underlying infrastructure evolves: VM families change, storage layers shift, operating systems are upgraded, kernels are updated, networking stacks evolve, and managed platforms introduce new constraints.

As this churn accelerates, differences in system behavior can surface even when PostgreSQL itself remains largely unchanged. Drawing from real-world operational experience running PostgreSQL in large-scale environments, this talk shares a few practical examples that illustrate how infrastructure changes surface as differences in database behavior.

These examples highlight how changes outside the database can influence behavior in ways that are not always immediately obvious.