Re: Correct way to accuse BCHR tuning method (Was: Hit ratio)

[Mogens Nørgaard]

I think Dave Ensor, at the recent UKOUG conference, called it his 
"portable tuning kit":

select elapsed_time, cpu_time
 from v$sql
order by elapsed_time;
Mogens

Connor McDonald wrote:

Yep.

Simple example: Even though it seems to be sometimes a
little on the 'random' side, the ELAPSED_TIME column
on V$SQL in v9 is an absolute god send...
Cheers
Connor
--- Mogens_Nørgaard <[EMAIL PROTECTED]> wrote: > And I
think it's important to realise that ratios
 

are useless as a 
starting point in the tuning process on any system,
not only Oracle. 
Most OS'es and databases use not instrumented
correctly to deal with 
response time measurements (makes you wonder: If
response TIME is what 
matters, how can you then not measure exactly that -
time?) - so in the 
other worlds (Unix, VMS, Windows, SQL Server, MySQL,
DB2....) people 
still beleive in the ratios because they have
nothing better.

Mogens

Anjo Kolk wrote:

   

BCHR tuning is useless as a starting point in the
     

tuning process. 
   

Anjo.

-----Original Message-----
Yong Huang
Sent: Wednesday, December 24, 2003 6:09 PM
To: Multiple recipients of list ORACLE-L
[This message is not technical, but educational.
     

Readers interested in
   

technical info only may want to skip]

Hi, Cary and Gopal,

My last message is misunderstood. Nowadays most
     

DBAs that still use
   

buffer cache hit ratio as a primary performance
     

tuning method are those
   

that rarely browse public forums. When we convince
     

them that's a wrong
   

method, we should not say "Look. I can bump up BCHR
     

to an arbitrary
   

value". If he doesn't think, he'll say "Indeed. If
     

I can get any value,
   

it must be rubbish". But if he's a logical person
     

and thinks for a few
   

minutes, he'll say "It's unfair to run that
     

choose_a_hit_ratio program
   

to get an arbitrary hit ratio and say the method is
     

wrong, because you
   

can use the same logic to write a program to get an
     

arbitrary library
   

cache hit ratio, OS in-core inode cache hit ratio
     

or directory name
   

cache hit..."

My last message is not meant to revive the outdated
     

and probably never
   

correct tuning method. Instead it's meant to let
     

oracle-l members know
   

that when you need to convince those DBAs that
     

still use that method,
   

you need to accuse the BCHR method for correct
     

reason, namely, BCHR does
   

not contain sufficient information for tuning, not
     

because you can raise
   

its value by constantly scanning a table in Oracle;
     

you won't be able to
   

convince some stubbon DBAs who enjoy thinking in a
     

quiet place.
   

I agree that "It's not the ratio that needs
     

condemning, it's the advice
   

about..." What I disagree is the wrong educational
     

tool people on public
   

forums have recently used again and again to show
     

the inadequacy of the
   

BCHR tuning method.

Yong Huang

__________________________________
Do you Yahoo!?
New Yahoo! Photos - easier uploading and sharing.
http://photos.yahoo.com/
     

--
Please see the official ORACLE-L FAQ:
http://www.orafaq.net
--
Author: =?ISO-8859-1?Q?Mogens_N=F8rgaard?=
 INET: [EMAIL PROTECTED]
Fat City Network Services    -- 858-538-5051
http://www.fatcity.com
San Diego, California        -- Mailing list and web
hosting services
   

---------------------------------------------------------------------
 

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). 
   

=====
Connor McDonald
web: http://www.oracledba.co.uk
web: http://www.oaktable.net
email: [EMAIL PROTECTED]
"GIVE a man a fish and he will eat for a day. But TEACH him how to fish, and...he will sit in a boat and drink beer all day"

________________________________________________________________________
Yahoo! Messenger - Communicate instantly..."Ping" 
your friends today! Download Messenger Now 
http://uk.messenger.yahoo.com/download/index.html
 

--
Please see the official ORACLE-L FAQ: http://www.orafaq.net
--
Author: =?ISO-8859-1?Q?Mogens_N=F8rgaard?=
 INET: [EMAIL PROTECTED]
Fat City Network Services    -- 858-538-5051 http://www.fatcity.com
San Diego, California        -- Mailing list and web hosting services
---------------------------------------------------------------------
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).