A classic example :

I got an e-mail this morning asking me to recreate a table with bigger
initial and second extents because they have bad response times. They talked
to the DBA ?? and he told them that it is because the table have many
extents. Checked the table. It has a lot of extents - 6.

This thread helped me a lot in telling them to go bark up another tree.

Yechiel Adar
Mehish
----- Original Message -----
To: Multiple recipients of list ORACLE-L <[EMAIL PROTECTED]>
Sent: Tuesday, September 10, 2002 11:16 PM


> Jared,
>
>     What version of the database does this book refer to?  I do remember
back in
> Version 6 and early 7 that having all of your data for a table and/or
index all
> in the first extent was a performance benefit.  Regrettably that idea had
been
> proved false more than once after 7.2 hit the street.  I notice a
performance
> improvement if tables/indexes are in more than one extent.  But if she's
talking
> about migrated and/or chained rows that's another matter & one that I'm
chasing
> as I type.
>
>     Looks like you've got a classic case of someone with enough
information to
> be dangerous.
>
> Dick Goulet
>
> ____________________Reply Separator____________________
> Author: [EMAIL PROTECTED]
> Date:       9/10/2002 12:28 PM
>
> <RANT>
>
> I've just spent 30 minutes with our SAP administrator trying to
> convince her that we really don't need to reorganize the tables
> in our production SAP database.
>
> Due to some misinformation in an Oracle Press book, 'Oracle Unleashed'
> I think, she is equating number of extents with fragmentation.
>
> The text she referred me to is in fact discussing 'migrated rows' though
> that term is never used.  She has become convinced that if the
> extents allocated for tables are not all in contigous space, some
> very nasty fragmentation will occur.
>
> I tried taking it down to disk and explaining that an OLTP system with
> hundreds of users won't really see much benefit from this, but she
> wasn't really ready for that.  :)
>
> Her concern is that there are 29000 extents in an index tablespace.
> This might have something to do with there being 3400 indexes in
> said tablespace.
>
> Total 'wasted' ( honeycomb ) space in this 250 gig DB is < 20 meg.  Not
> much to  gain there.
>
> The text of the book states that you should expect a '10 to 20 percent
> performance increase' by reorganizing the tables/indexes.  No data to
> back it up of course.
>
> This is on a database that performs very well most of the time, outside
> of a couple of custom reports that run too long.  No complaints from
> users about slowness.
>
> Arrghhh!
>
> I just had to vent to the list, cuz there's no one here that understands.
>
> <\RANT>
>
> Jared
>
> --
> Please see the official ORACLE-L FAQ: http://www.orafaq.com
> --
> Author:
>   INET: [EMAIL PROTECTED]
>
> Fat City Network Services    -- (858) 538-5051  FAX: (858) 538-5051
> San Diego, California        -- Public Internet access / Mailing Lists
> --------------------------------------------------------------------
> To REMOVE yourself from this mailing list, send an E-Mail message
> to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in
> the message BODY, include a line containing: UNSUB ORACLE-L
> (or the name of mailing list you want to be removed from).  You may
> also send the HELP command for other information (like subscribing).
> --
> Please see the official ORACLE-L FAQ: http://www.orafaq.com
> --
> Author:
>   INET: [EMAIL PROTECTED]
>
> Fat City Network Services    -- (858) 538-5051  FAX: (858) 538-5051
> San Diego, California        -- Public Internet access / Mailing Lists
> --------------------------------------------------------------------
> To REMOVE yourself from this mailing list, send an E-Mail message
> to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in
> the message BODY, include a line containing: UNSUB ORACLE-L
> (or the name of mailing list you want to be removed from).  You may
> also send the HELP command for other information (like subscribing).

-- 
Please see the official ORACLE-L FAQ: http://www.orafaq.com
-- 
Author: Yechiel Adar
  INET: [EMAIL PROTECTED]

Fat City Network Services    -- (858) 538-5051  FAX: (858) 538-5051
San Diego, California        -- Public Internet access / Mailing Lists
--------------------------------------------------------------------
To REMOVE yourself from this mailing list, send an E-Mail message
to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB ORACLE-L
(or the name of mailing list you want to be removed from).  You may
also send the HELP command for other information (like subscribing).

Reply via email to