From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | "Matthew Churcher" <Matthew(dot)Churcher(at)realvnc(dot)com> |
Cc: | "'Thom Brown'" <thom(at)linux(dot)com>, pgsql-general(at)postgresql(dot)org |
Subject: | Re: Having trouble with pg_dumpall -o |
Date: | 2012-05-01 13:35:18 |
Message-ID: | 11724.1335879318@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
"Matthew Churcher" <Matthew(dot)Churcher(at)realvnc(dot)com> writes:
> The triggers are being used to track changes to the tables. The developers
> are concerned that using string references for the table names in this case
> would create too much overhead as this is a frequent operation and is
> performance critical.
"Premature optimization is the root of all evil". Do you have any
actual evidence that you bought a meaningful amount of performance
with this idea? It seems like a terrible restriction from here
--- for instance, you are absolutely locked out of doing plain
dump and restore.
> Sounds like we have the choice of using string names or implementing a
> custom migration script.
pg_upgrade might save your bacon here. But I'd rewrite that code ASAP.
The fact that pg_upgrade preserves table OIDs is an implementation
detail, not a feature we promise to preserve.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Adrian Klaver | 2012-05-01 13:38:44 | Re: Having trouble with pg_dumpall -o |
Previous Message | Schade, Jeffrey | 2012-05-01 12:58:38 | Table / View Security Report |