| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | "Albe Laurenz" <all(at)adv(dot)magwien(dot)gv(dot)at> |
| Cc: | "Richard Broersma Jr *EXTERN*" <rabroersma(at)yahoo(dot)com>, rongkai(dot)zhao(at)gmail(dot)com, pgsql-general(at)postgresql(dot)org |
| Subject: | Re: Rule vs Trigger |
| Date: | 2007-06-26 14:04:26 |
| Message-ID: | 18188.1182866666@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
"Albe Laurenz" <all(at)adv(dot)magwien(dot)gv(dot)at> writes:
> Richard Broersma Jr wrote:
>> My understanding is that Triggers offer better performance
>> than rules do.
> A trigger FOR EACH STATEMENT will execute the trigger function
> for each row affacted by the statement.
Huh? That would be true for a FOR EACH ROW trigger, but a STATEMENT
trigger fires once per statement.
> A rule would only
> execute one additional statement. So if you can do it with a rule
> conveniently, the rule will probably be faster.
I find this unlikely. The overhead involved in setting up a rule
is probably larger than that involved in calling a trigger.
The real question is whether you need access to the modified data
or not --- a statement-level trigger doesn't currently get that.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Alvaro Herrera | 2007-06-26 14:21:42 | Re: LC_CTYPE and matching accented chars |
| Previous Message | Gurjeet Singh | 2007-06-26 13:44:51 | Re: a JOIN on same table, but 'slided over' |