Re: ** high water mark for small tables

From: Jared Still <jkstill_at_gmail.com>
Date: Mon, 27 Jul 2009 16:54:05 -0700
Message-ID: <bf46380907271654g4db93a9fof5da4b324d86ce12_at_mail.gmail.com>

On Mon, Jul 27, 2009 at 3:06 PM, Bobak, Mark <Mark.Bobak_at_proquest.com>wrote:

> Agreed. If you have to hit most of the table blocks, you can’t beat an
> FTS. What I think Jared was getting at, is that there is (or at least used
> to be) a myth that small tables did not have to be indexed. It can easily
> be demonstrated that an index can be beneficial, even for a single-row
> table.
>
>

Exactly.

Also, please notice that I did not say 'must implement'.

My post said to 'give serious consideration to'.

eg. look at the access path, determine if an index is beneficial, test it, and implement it if it works.

Jared

--
http://www.freelists.org/webpage/oracle-l

Received on Mon Jul 27 2009 - 18:54:05 CDT