| From: | Alexander Korotkov <a(dot)korotkov(at)postgrespro(dot)ru> |
|---|---|
| To: | Claudio Freire <klaussfreire(at)gmail(dot)com> |
| Cc: | Peter Geoghegan <pg(at)bowt(dot)ie>, Simon Riggs <simon(at)2ndquadrant(dot)com>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: Locking B-tree leafs immediately in exclusive mode |
| Date: | 2018-06-14 14:13:14 |
| Message-ID: | CAPpHfdufVgOfpG+ZoujrrFQQYMfuJseAC9Ypi7Sp3SjnHvQ+XA@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Thu, Jun 14, 2018 at 4:56 PM Claudio Freire <klaussfreire(at)gmail(dot)com> wrote:
> Not at all. Insertion cost in unique indexes with lots of duplicates
> (happens, dead duplicates) grows quadratically on the number of
> duplicates, and that's improved by making the index unique and sorted.
Sorry, I've messed up the terms. I did actually compare current
non-unique indexes with non-unique indexes keeping duplicate entries
ordered by TID (which makes them somewhat unique). I didn't really
considered indexes, which forces unique constraints. For them
insertion cost grows quadratically (as you mentioned) independently on
whether we're keeping duplicates ordered by TID or not.
------
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Alvaro Herrera | 2018-06-14 14:23:28 | Re: Needless additional partition check in INSERT? |
| Previous Message | Claudio Freire | 2018-06-14 13:56:17 | Re: Locking B-tree leafs immediately in exclusive mode |