Exploring PostgreSQL 18's new UUIDv7 support

Re:Exploring PostgreSQL 18's new UUIDv7 support by grapesodaaaaa on Friday October 17, 2025 @10:41PM (#45624457)

I still feel like calling something like uuid.v4() is easier and less cognitively complex.

[ Parent | on HN ]

Re:Exploring PostgreSQL 18's new UUIDv7 support by cyberax on Friday October 17, 2025 @11:22PM (#45624643)

There are advantages in monotonically increasing UUIDs, they work better with BTrees and relational databases.

[ Parent | on HN ]

Re:Exploring PostgreSQL 18's new UUIDv7 support by grapesodaaaaa on Saturday October 18, 2025 @11:58AM (#45628336)

I just meant having UUIDv7 internally, and UUIDv4 externally if date leakage is a concern (both on the same object).

UUIDv7 still works great in distributed systems and has algorithmic advantages as you have mentioned.

[ Parent | on HN ]