23 Times Faster Inserts With UUIDv7 | Scaling Postgres 432

In this episode of Scaling Postgres, we discuss getting 23 times faster inserts with UUIDv7, the benefits and downsides of UUIDs, how to optimize queries without table changes and using WAIT FOR to read your writes.

Content Discussed

• PostgreSQL 18: 23x Faster Inserts With UUID V7
• The Time Traveler's Primary Key
• How to optimize when you can’t do anything!
• Read your writes: WAIT FOR in PostgreSQL 19
• Estimating work_mem
• The history of Postgres sharding
• Problems with large tables in Postgres
• New things for regular expressions in PostgreSQL (pg_tre and pg_re2)
• Day to day PostgreSQL on Kubernetes: Resize, Scale, and Upgrade
• pgwatch v6: Prometheus becomes a source, not just a sink
• Do Global Hash Tables Strike Back in PostgreSQL?
• The Sixth Execution
• When huge_pages Help: Using PSI to Find Bottlenecks Before You Decide
• How to Tell Whether Your Slow Query Is a Planning or an Execution Problem
• The Postgres Insert That Fails Right After a Successful Load
• MongoDB on PostgreSQL: DocumentDB with pglayers-azure
• The Request Becomes a Transaction#
• Scenario-Tree Testing in PostgreSQL: Every Authored Branch, Shared History, Before COMMIT#
• Your PostgreSQL Platform Has Telemetry. Does It Have a Digital Twin?
• Contributions for week 33
• PostgreSQL in Taipei: Connecting Taiwan to the Global PostgreSQL Community (COSCUP 2026)
• pg_statviz 1.2 released with PostgreSQL 19 support and new features
• Enterprise-managed auth for the Supabase MCP server

YouTube Video

Podcast Audio