Re: Sort on column def.

From: Tanel Poder <tanel_at_poderc.com>
Date: Wed, 9 Sep 2009 23:19:51 +0800
Message-ID: <4602f23c0909090819l62b78007lbd60ef8384439ee1_at_mail.gmail.com>

Not sure whether it answers your question directly, but Jonathan Lewis has written about sorted hash cluster internals in past:

http://jonathanlewis.wordpress.com/2008/07/13/sorted-hash-clusters/

-- 
Tanel Poder
http://blog.tanelpoder.com


On Wed, Sep 9, 2009 at 10:56 PM, Brady, Mark
<Mark.Brady_at_constellation.com>wrote:



>  *10gR2 Docs:*

> SORT

> The SORT keyword is valid only if you are creating this table as part of a

> hash cluster and only for columns that are also cluster columns.

> This clause instructs the database to sort the rows of the cluster on this

> column before applying the hash function. Doing so may improve response time

> during subsequent operations on the clustered data.

>

>

> I don’t get it. My rudimentary understanding is: Oracle uses a hash

> function on the clustered column that would produce a “Block offset” and

> place that row there… so the location of the row is predetermined, what does

> it buy to sort the inserts? Is it purely an efficient space allocation

> thing? If done in order, the less likely to chain rows? Something like that?

>

>

>


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

Received on Wed Sep 09 2009 - 10:19:51 CDT