Hacker News new | ask | show | jobs
by wslh 1113 days ago
Basic question: is it correct to assume that having PGQ involves a big change in the database engine?
2 comments

AFAICT the idea is that you are not directly querying the tables as a graph, but you construct a graph "view" from the tables, and then query that graph using PCQ.
Correct. The one RD I know that has implemented this "graph view" concept is Oracle's. They did it first with PGQL extensions on top of say 19c. Interesting, but the execution seemed a bit unpolished.

Now with 23c, they are adopting PGQ and one should be able to interface through the regular JDBC connectors. Will see how it shapes up.

It's just a different language and a simple "property" layer over the existing data. No changes to the internals are necessary.
No changes are required by PGQ but some engines may wind up using PGQ as a signal to bring in new types of indexes or disk storage formats to back it.

It may be interesting to see if real world usage of PGQ pushes the database engines to do interesting things.