To get the show notes as well as get notified of new episodes, visit:
https://www.scalingpostgres.com/episodes/394-five-times-faster-aggregates/

In this episode of Scaling Postgres, we discuss enhancements bringing more than five times faster aggregates and 64% faster counts, better performance through sampling and my course sale is ending.

Want to learn more about Postgres performance? 
Join my FREE training called Postgres Performance Demystified here: https://www.scalingpostgres.com/courses/postgres-performance-demystified/

Timestamps:
00:10 - Ludicrous Speed Postgres Sale Ends 12/1
00:57 - Super fast aggregations in PostgreSQL 19
03:55 - Settling COUNT(*) vs COUNT(1) debate in Postgres 19
05:54 - PostGIS Performance: Data Sampling
08:12 - What do the new Index Searches lines in EXPLAIN mean?
09:01 - Using Explain
13:23 - What's new in EXPLAIN
13:48 - We Taught AI to Write Real Postgres Code (And Open Sourced It)
15:37 - AI-Powered Postgres index suggestions
15:57 - Just Use Postgres!
16:10 - Outro

#postgres #postgresql

Scaling Postgres

A roundup of PostgreSQL performance news covering several topics: PostgreSQL 19 will introduce aggregate-before-join optimization yielding up to 5x faster aggregate queries; PostgreSQL 19 also improves COUNT(*) vs COUNT(1) vs COUNT(column) performance with up to 64% gains for NOT NULL columns. Additional topics include using TABLESAMPLE for statistical sampling on large tables (including geospatial data), understanding the new index search count lines in EXPLAIN output introduced in PostgreSQL 18, skip scan behavior and multi-column index ordering strategies, an open-sourced MCP server for generating better PostgreSQL schemas via LLMs, and AI-powered index suggestions from PlanetScale.

5 Times Faster Aggregates | Scaling Postgres 394