Re: Oracle slowed down

On Fri, 21 Apr 2006 08:33:05 -0700, stevedhoward wrote:

> I think I understand where Bob is coming from. I think he is saying that
> *after* you have tuned the application, you could look to get more of the
> remaining required buffers from Oracle memory than from disk (whatever
> that is nowadays, with OS cache, SAN cache, etc.). One logical way (once
> again, after you have removed the low hanging SQL fruit) is to "encourage"
> Oracle to get more blocks that are needed from memory rather than "disk".
>

The one issue I have with BHCR is simple - people have gotten into the habit of using it as the primary tuning tool.

However, makes a huge assumption ... it ONLY gives us information about whether we are finding stuff in the cache - again. Useless if noone else is interested in the same information,

A 100% BCHR can just as easily identify a grossly overconfigured cache as it can identify popular data sets. In both cases the misinterpretation is that it looks good.

-- 
Hans Forbrich                           
Canada-wide Oracle training and consulting
mailto: Fuzzy.GreyBeard_at_gmail.com   
*** Top posting [replies] guarantees I won't respond. ***