Schedule - PostgreSQL Development Conference 2024

Collations from A to Z

Date: 2024-05-29
Time: 14:30–15:20
Room: Canfor(1600)
Level: Intermediate
Feedback: Leave feedback

Comparisons are fundamental to computing, and “putting words in order” is not as straightforward as many people initially think. Collations, or the ordering and comparison of strings, is continuously evolving along with all the nuances of natural language. But databases use such comparisons everywhere: for ORDER BY, the humble >, <, and = operators, btree indexes, GROUP BY, range partitioning -- even hashing, hash indexes, and hash partitioning. Fundamental decisions from the early days of PostgreSQL have led to surprising challenges. Learn how to navigate these challenges and what new options are available.

In this session, we will discuss:

  • a brief overview and history of collation options Postgres offers
  • the implications of those options on performance, integrity, and usability of Postgres
  • how to configure the database collation, and when to use the COLLATE clause
  • case-insensitive matching and advanced tricks
  • mitigating integrity risks, especially around indexes
  • upcoming developments / ongoing work

Talk recorded at


The following slides have been made available for this session:


Jeff Davis
Jeremy Schneider