Patrick Scharrenberg, 04.08.2008 17:51:
> I have to do much inserts into a database where the key most often is
> already there.
> My current approach is to query for the key (ip-address), and if the
> result is null I do the insert.
> For every IP-Address I need the ip_addr_id from the same table.
> Now I'm wondering if there is a better solution, since I'm doing ~20
> inserts at once and every time I'm doing single lookup's for the IDs.
If you know that most of the time the record is already there, I simply execute the UPDATE, then check how many rows were updated. If that returns zero, I'll send the INSERT
In response to
pgsql-sql by date
|Next:||From: Richard Broersma||Date: 2008-08-04 15:59:58|
|Subject: Re: fast insert-if-key-not-already-there|
|Previous:||From: Patrick Scharrenberg||Date: 2008-08-04 15:51:27|
|Subject: fast insert-if-key-not-already-there|