Title: RE: Anyone seen any independent performance .....

There’s actually more to it than that. For every block in Oracle that changes, there’s:

 

1. Generation of redo on the upcoming undo block change

2. An undo (i.e., rollback segment) block change

3. Generation of redo on the upcoming data block change

4. A data block change

 

These four steps happen for every index and data block change. So, inserting into a table with n indexes will motivate on the order of (1+n)*2 units of redo to be written to your redo stream, and (1+n)*(1+1) blocks of undo and data to be written through your DBWR (actually more, because index leaf node insertions sometimes generate recursive SQL that propagates changes upward in the B*-tree, each block of which also generates redo and undo).

 

Cary Millsap
Hotsos Enterprises, Ltd.
http://www.hotsos.com

Upcoming events:
- Hotsos Clinic, Oct 1–3 San Francisco, Oct 15–17 Dallas, Dec 9–11 Honolulu
- 2003 Hotsos Symposium on Oracle® System Performance, Feb 9–12 Dallas
- Next event: Miracle Database Forum, Sep 20–22 Middlefart Denmark

-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] On Behalf Of Steve Perry
Sent: Thursday, September 05, 2002 1:53 PM
To: Multiple recipients of list ORACLE-L
Subject: Re: Anyone seen any independent performance .....

 

(1) one for the table insert.

+

(3) 1 for the header block, 1 for the branch block and 1 for the leaf block * number of indexes

 

I think Dave Ensor mentioned something like that in a presentation.

 

----- Original Message -----

Sent: Thursday, September 05, 2002 1:21 PM

Subject: RE: Anyone seen any independent performance .....

 

I don't remember where but the results of the study were like following ...

if the cost of inserting one row to a table is 1 unit and if you have 5 indexes on the table than total cost of inserting 1 row to the table is

     1   (cost of inserting a row)
 + 3*5 (5 indexes)
---------
   16  

So total cost is about 16 units.

Raj
______________________________________________________
Rajendra Jamadagni              MIS, ESPN Inc.
Rajendra dot Jamadagni at ESPN dot com
Any opinion expressed here is personal and doesn't reflect that of ESPN Inc.
QOTD: Any clod can have facts, but having an opinion is an art!

 

-----Original Message-----
From: Johnson, Michael [mailto:[EMAIL PROTECTED]]
Sent: Thursday, September 05, 2002 2:08 PM
To: Multiple recipients of list ORACLE-L
Subject: Anyone seen any independent performance .....

 

studies on what happens to performance as indexes keep getting added to tables while inserting data ??

Reply via email to