Re: Oracle database configuration options

From: Brian M. Biggs <bbiggs_at_cincom.com>
Date: 1996/11/01
Message-ID: <327A4EF9.63B3_at_cincom.com>#1/1


Thomas J. Kyte wrote:
>
> >>So we thought 'OK, let's just use VARCHAR2 all the time', but were
> >>advised that this wasn't a good idea, as there would be a significant
> >>performance impact when using VARCHAR2 columns in indexes, so we didn't
> >>bother.
> >>We're not yet too far down the road to change our approach, and would
> >>welcome any comments or suggestions on this.
> >
> >I don't know of any formal trials, but the general view seems to be that
> >there is no significant performance hit when using VARCHAR2
> >--
> >Jim Smith
>
> there are none, a CHAR field is really a VARCHAR2 with blanks to pad it out to
> the max size. therefore a CHAR takes up the same amount of space (leading size
> indicator followed by data, the size is constant for the entire column, blanks
> pad out).

As I commented in the other thread on this topic, 'A question of principle', we HAVE seen a negative performance impact when switching from CHAR to VARCHAR2 datatype. We are using Pro*COBOL, and I am guessing the problem stems from the fact that COBOL likes to have everything padded with spaces, and the time increase we are seeing may be the Oracle interface padding for COBOL and unpadding for Oracle the VARCHAR2 fields as they go back and forth to and from the database.

This also isn't to say that we aren't implementing incorrectly. We are using Pro*COBOL and compiling in ANSI mode, which may be another issue. How important is it that we remain ANSI compliant for going to other databases (God forbid we use something other than Oracle!!!)? There are important differences in the way the Oracle interface behaves, at least in Pro*COBOL, depending on whether you compile in ANSI or Oracle mode.

Regards,
Brian

-- 
Brian M. Biggs				mailto:bbiggs_at_cincom.com
Cincom Systems, Inc.			voice: (513) 677-7661
http://www.cincom.com/
Received on Fri Nov 01 1996 - 00:00:00 CET

Original text of this message