Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: Design approaches about primary key

Re: Design approaches about primary key

From: Sybrand Bakker <gooiditweg_at_sybrandb.nospam.demon.nl>
Date: Sat, 29 Nov 2003 19:57:38 +0100
Message-ID: <dtqhsvcqlrpnogdfso8hn90m2spvbpe3pa@4ax.com>


On Fri, 28 Nov 2003 22:52:00 +0100, "Sven Kolar" <devnull_at_svenkolar.net> wrote:

>OLTP:
>Surrogate keys should IMHO be avoided in OLTP
>implementations. Main reason: avoiding excessive lookups
>during atomic-transaction SELECTs, UPDATEs and
>INSERTs (which are plentiful in a typical OLTP setup). This
>approach simplifies my access logic (client-side DML), plus
>lessens the load on the server (no lookup joins!) -- join-hashing
>notwithstanding.
>

Utter crap. Lookup tables should be in the keep buffer pool. And lookup tables are usually small.

--
Sybrand Bakker, Senior Oracle DBA
Received on Sat Nov 29 2003 - 12:57:38 CST

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US