From: | pabloa98 <pabloa98(at)gmail(dot)com> |
---|---|
To: | Adrian Klaver <adrian(dot)klaver(at)aklaver(dot)com> |
Cc: | Rob Sargent <robjsargent(at)gmail(dot)com>, "Peter J(dot) Holzer" <hjp-pgsql(at)hjp(dot)at>, "pgsql-generallists(dot)postgresql(dot)org" <pgsql-general(at)lists(dot)postgresql(dot)org> |
Subject: | Re: Could postgres12 support millions of sequences? (like 10 million) |
Date: | 2020-03-23 00:35:33 |
Message-ID: | CAEjudX7VxHjpv3i8e+evtwVSdTWJce7j6aK32a7fNxkNKE-bWA@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
> > Now I read this paragraph, I realize I was not clear enough.
> > I am saying we do not want to use locks because of all the problems
> > described.
>
> And what I was asking is what locking where you doing?
>
> And it might be better to ask the list how to solve those problems, then
> to create a whole new set of problems by using millions of sequences.
>
> What we are doing in the **OTHER section** with row locks is basically:
BEGIN;
...
x = SELECT last_value + 1 FROM counter WHERE group = ... AND element = ....
FOR UPDATE;
...
if everything is fine, then:
UPDATE counter SET last_value = x WHERE group = ... AND element = ....
FOR UPDATE;
COMMIT;
else:
ROLLBACK;
From | Date | Subject | |
---|---|---|---|
Next Message | Christopher Browne | 2020-03-23 00:36:25 | Re: Could postgres12 support millions of sequences? (like 10 million) |
Previous Message | Adrian Klaver | 2020-03-23 00:10:10 | Re: Duplicate key violation on upsert |