Message-ID: <F001.005D10B6.20030925050942@fatcity.com>
Date: Thu, 25 Sep 2003 05:09:42 -0800
To: Multiple recipients of list ORACLE-L <ORACLE-L@fatcity.com>
Sender: ml-errors@fatcity.com
Reply-To: ORACLE-L@fatcity.com
Errors-To: ML-ERRORS@fatcity.com
From: "Mark Leith" <mark@cool-tools.co.uk>
Subject: RE: guidance
Organization: Fat City Network Services, San Diego, California
Precedence: bulk
Mime-Version: 1.0
Content-Type: text/plain;	charset="Windows-1252"
Content-Transfer-Encoding: 7bit

http://www.theregister.co.uk/content/53/33042.html


-----Original Message-----
Stephane Faroult
Sent: 25 September 2003 02:45
To: Multiple recipients of list ORACLE-L


<RANT>
Well, concerning point 4), I am surprised by the resilience of users to
often dreadful applications. Perhaps that with age I am getting more and
more impatient, but in their place I would have flown terminal and
keyboard across the room. Perhaps I have memories of a time when
machines weren't even a shadow of today's, and performance were hardly
worse than many things you see running today. When you try to measure
the amount of 'business units processed per unit of work', it is
pathetic more often than it should be. I guess that what saves many
designers and developers from being lynched by popular justice is that
most users have no idea about what *could* be done and are ready to
swallow that Oracle is slow, their 64 processor machine not powerful
enough, etc, etc.
</RANT>

SF


Tanel Poder wrote:
>
> Forget the modern tuning skills and when you're asked what shows the best
> that your database works optimally;
>
> 1) Buffer cache hit ratio is 99%
> 2) Buffer cache hit ratio is 99,999%
> 3) Buffer cache hit ratio is 999999%
> 4) Users aren't complaining
>
> Then answer 3 for sure ;)
>
> Tanel.
>
> ----- Original Message -----
> To: "Multiple recipients of list ORACLE-L" <ORACLE-L@fatcity.com>
> Sent: Thursday, September 25, 2003 12:04 AM
>
> > List , I am planning to give my 9i performance tuning exam on the first