| From: | Merlin Moncure <mmoncure(at)gmail(dot)com> |
|---|---|
| To: | Ivan Voras <ivoras(at)freebsd(dot)org> |
| Cc: | Moshe Jacobson <moshe(at)neadwerx(dot)com>, pgsql-general(at)postgresql(dot)org |
| Subject: | Re: Re: What's faster? BEGIN ... EXCEPTION or CREATE TEMP TABLE IF NOT EXISTS? |
| Date: | 2012-10-05 13:55:18 |
| Message-ID: | CAHyXU0zVRmYP+6j6R5Q960sowodsGRzX+ZPtWRo6nfZWbTuK-w@mail.gmail.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
On Fri, Oct 5, 2012 at 3:09 AM, Ivan Voras <ivoras(at)freebsd(dot)org> wrote:
> I think I can make a fairly educated guess that catching exceptions
> while dealing with session variables should be much, much faster than
> creating any kind of a table :)
That is true, but it's not clear how using session variables keeps you
from having to create the table. If the table is already there, a
session variable guarding the table construction shouldn't be any
faster/better than a simple 'create if not exists'. A catalog scan is
basically boils down to a query. So I'm not sure that actually helps.
On the other hand, if the temp table can be completely ditched for a
session variable or two, then yeah, that would be much better since
you'd avoid the overhead of creating the table completely.
merlin
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Moshe Jacobson | 2012-10-05 14:29:54 | Counting all results before LIMIT |
| Previous Message | Shaun Thomas | 2012-10-05 13:43:17 | Re: Trajectory of a [Pg] DBA |