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

Home -> Community -> Usenet -> c.d.o.tools -> Re: Using a Number datatype

Re: Using a Number datatype

From: Mark D Powell <markp7832_at_my-deja.com>
Date: 2000/05/18
Message-ID: <8g0rfh$1tu$1@nnrp1.deja.com>#1/1

In article <3922E60F.14F62C55_at_motorola.com>,   Allen <allen.brost_at_motorola.com> wrote:
> What is the default size of a Number datatype if no value
> is given when the table is created? If you know your going
> to to use a number below 100, does it offer a performance
> benefit both in speed and space to define the column as
> Number(2)?
>
> - Allen
>

Defining a column that will hold a number in the range of 0 - 99 as number(2) vs number has no storage space or real performance benefit, but it does allow the rdbms to enforce value checking for the domain. I would say it is the prefered form as you want as much validity checking as possible on the data. Defining a number column as only 2 digits will result in less space being necessary to display/handle the output and there may be some small performance consideration here but that gets into the real internals that few people can speak about with authority.

Oracle stores all numberic datatypes internally as type number which is a form of scientific notation stored in variable length fields of no more than 22 bytes. The formula is in the Concepts manual and is something like 1 + (floor(digits/2) + 1) + 1, this last +1 is for the sign and is only required if the value is negative.

--
Mark D. Powell  -- The only advice that counts is the advice that
 you follow so follow your own advice --


Sent via Deja.com http://www.deja.com/
Before you buy.
Received on Thu May 18 2000 - 00:00:00 CDT

Original text of this message

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