RE: pctused bs

Amen, my straight-talking brother!

<sarcasm>
But, why on Earth don't you want to spend hours spewing arcane 0.001% performance tweaks you learned during the Advanced OCP Certification Boot Camp?

Don't you want to impress people? Don't you want a limo, yer own helicopter, and an oracle web site? Come on Greg, get with the program, just working hard, understanding internals, and keeping the customer's database up and running well is not enough. You need also to acquire the title of guru, genius, wizard, etc.
<sarcasm/>

Thanks for the post, it was refreshing, much in the same way as a seeing Harrison Ford (or was it Croc Dundee?) just pull out a gun and shoot the 'martial arts expert' was.

I think anyone who advocates clever tuning tweaks is duty-bound to supply some hardcore supporting evidence. There are such people, but they are VERY few and far between. (Anjo Kolk, Cary Millsap, and Craig Shallahammer come immediately to mind. ) This is because it takes hard work to build a strong case for performance ( look at the work that goes into TPC, fer instance.)

Basically, when DBA tuning tips are peer-reviewed at the same level as, say, a scientific paper, then a fair bit of the "magic", covert behavior and ego will go out of it, and be replaced by repeatability, transparency, and better value for our customers. And careers.

Jeez, I didn't mean to be so long-winded and serious. I'll have to make alot of silly jokes to make up for it now.

|| -----Original Message-----
|| From: Greg Moore [mailto:sqlgreg_at_pacbell.net]
|| Sent: Thursday, June 07, 2001 3:00 AM
|| To: Multiple recipients of list ORACLE-L
|| Subject: pctused bs
||
||
|| The general theory about increasing PCTUSED is that blocks
|| will be fuller,
|| but they may go on and off the freelist more often and that
|| will hurt DML
|| performance.
||
|| But in reality what is the overhead of this? Isn't the need
|| to bring more
|| blocks in and out of the data buffer greater than the
|| overhead of taking
|| such a block on and off the freelist?
||
|| How the heck does a person measure what the overhead is of
|| taking a block on
|| and off the freelist? Furthermore, even if you have a
|| measure, how do you
|| compare the disadvantage from fuller blocks going on and off
|| a freelist
|| against the advantage of fuller blocks making full table
|| scans faster?
||
|| I see all this stuff in books, but it's just a lot of
|| general statements
|| that amount to so much garbage. It can sound fancy to talk
|| about PCTUSED,
|| and if you can do it in a book you can pose as an expert and
|| fool most
|| hiring managers into thinking you're an expert, but unless
|| you can actually
|| measure the gains and the losses from changing this
|| parameter, what good is
|| any of this talk?
||
|| --
|| Please see the official ORACLE-L FAQ: http://www.orafaq.com
|| --
|| Author: Greg Moore
|| INET: sqlgreg_at_pacbell.net
||
|| 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: ListGuru_at_fatcity.com (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: Mohan, Ross
  INET: MohanR_at_STARS-SMI.com

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: ListGuru_at_fatcity.com (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).