Date: 2025-04-25
Time: 10:10–10:55
Room: West
Level: Intermediate
The release announcement for Postgres 16 included this cryptic note:
Finally, this release begins adding support for bidirectional logical replication, introducing functionality to replicate data between two tables from different publishers.
What’s this? Does that mean we can just start building Active-Active Postgres write clusters? Neat!
Slow down there! Running a distributed Postgres cluster isn’t all sunshine and rainbows! You need to account for new sequence rules, data merging concerns, conflict management, and the dreaded potential of node divergence.
This talk will present the revised theoretical framework necessary to successfully operate a distributed Postgres cluster. We’ll tell you how to build and operate one, but also how to modify your tables and other critical preparation steps to make your cluster ready for worldwide distribution.
This talk will also cover Spock from pgEdge, a new Open Source extension that provides the tools to transform Postgres into a distributed cluster powerhouse. Simple bi-directional replication isn’t quite enough, but the beauty of Postgres extensibility makes it possible.