Re: individual query hit ratio

From: Niall Litchfield <n-litchfield_at_audit-commission.gov.uk>
Date: Fri, 28 Nov 2003 08:57:28 -0000
Message-ID: <3fc70dfa$0$9386$ed9e5944@reading.news.pipex.net>

"Robert" <grobert_at_azet.sk> wrote in message news:bq6ppa$l62$1_at_news.telecom.sk...
> fine anna ...
>
> I do not need any philosophical answer. My question is simple. Is the
above
> query suitable for measuring "individual query cache hit ratio" or not. So
> if you know the answer please say it and do not talk about the sun spots.
> The reason why I`m asking is that I`m not familiar with many of the Oracle
> dynamic views and your answer did not helpd me to uderstand them more :-)

Its a reasonable stab at it. There might be some stats for queries that do not show up in v$sqlarea for example sort direct writes I don't *think* is included in the disk stats.

What lies behind Anna's reply is the realization that the hit ratio is an (almost) useless measure however you calculate it. If it changes dramatically it is probably a clue that something has changed with your database - unfortunately the direction of change of the statistic doesn't tell you whether performance has improved or worsened.

-- 
Niall Litchfield
Oracle DBA
Audit Commission UK

Received on Fri Nov 28 2003 - 02:57:28 CST