Show Notes
Nik and Michael are joined by Harry Brundage from Gadget to talk about their recent zero-downtime major version upgrade, how they use Postgres more generally, their dream database, and some challenges of providing Postgres as an abstracted service at scale.
Here are some links to things they mentioned: Harry Brundage https://postgres.fm/people/harry-brundage Gadget https://gadget.dev Zero downtime Postgres upgrades using logical replication (blog post) https://gadget.dev/blog/zero-downtime-postgres-upgrades-using-logical-replication HOT updates https://www.postgresql.org/docs/current/storage-hot.html PgDog https://pgdog.dev Multigres https://multigres.com Neki https://www.neki.dev Running 10 Million PostgreSQL Indexes In Production (Heap blog post) https://www.heap.io/blog/running-10-million-postgresql-indexes-in-production pgwatch2 (Postgres.ai Edition) https://gitlab.com/postgres-ai/pgwatch2 Advanced query insights on AlloyDB https://cloud.google.com/alloydb/docs/advanced-query-insights-overview OrioleDB https://www.orioledb.com Not discussed but relevant: Gadget have also now published a blog post about their sharding! https://gadget.dev/blog/sharding-our-core-postgres-database-without-any-downtime
~~~ What did you like or not like? What should we discuss next time? Let us know via a YouTube comment, on social media, or by commenting on our Google doc!
~~~ Postgres FM is produced by: Michael Christofides, founder of pgMustard Nikolay Samokhvalov, founder of Postgres.ai
With credit to: Jessie Draws for the elephant artwork
More from Postgres FM
Nik and Michael are joined by Álvaro Hernández to discuss sharding Postgres with Citus and StackGres, especially the new query router support that has been added very recently.
Nik and Michael are joined by Salma El-Sayed and Kirk Wolak to discuss their in-progress and very ambitious Google Summer of Code project around adding Btree Merge support to Postgres.
Nik and Michael are joined by Sai Srirampur to discuss ClickHouse's new managed Postgres service.
Nik and Michael discuss hot_standby_feedback — what it controls, whether and when read replicas should be changed to on, and other settings you can change to help too.
Nik and Michael are joined by Shaun Thomas to discuss estimating work_mem, memory management in general, and writing an extension to help.