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: Importing huge data

Re: Importing huge data

From: T. Bone <tommy914_at_hotmail.com>
Date: 5 Sep 2003 08:01:36 -0700
Message-ID: <fe01cfb5.0309050701.203893a1@posting.google.com>


"Howard J. Rogers" <howardjr2000_at_yahoo.com.au> wrote in message news:<3f550339$0$14559$afc38c87_at_news.optusnet.com.au>...
> The direct path insert is a modified form of the insert statement. Instead
> of it reading 'insert into X values ('A','B','C')', it reads 'insert
> /*+APPEND */ into X values ('A",'B','C')'. The append hint there causes data
> blocks to be constructed entirely in memory, and then slammed down on disk
> in a way that's guaranteed not to interfere with data that's already in the
> table. By doing the bulk of its work in memory, it's faster than a normal
> insert. And by respecting the nologging keyword, it has less overhead than a
> regular insert, and is thus faster again.

Howard,
The Oracle docs state that direct path insert only works when using the format Insert /*+APPEND */ into ... Select ... from syntax. Here is the reference from Oracle:
http://download-west.oracle.com/docs/cd/B10501_01/server.920/a96524/c21dlins.htm#11358

"Note: Direct-path INSERT supports only the sub query syntax of the INSERT statement, not the VALUES clause. For more information on the subquery syntax of INSERT statements, see Oracle9i SQL Reference ".

Is that not correct?

Tommy Received on Fri Sep 05 2003 - 10:01:36 CDT

Original text of this message

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