You're right. Postgres doesn't give any control over primary data locality. That might cause querying 1 row a bit faster in Postgres (no log(N) traversal of the

You're right. Postgres doesn't give any control over primary data locality. That might cause querying 1 row a bit faster in Postgres (no log(N) traversal of the primary B-tree index) but picking out N rows could be a lot slower. https://www.postgresql.org/docs/13/indexes-index-only-scans.html https://www.postgresql.org/docs/13/indexes-index-only-scans....