Re: High elapsed/cpu ratio

On Sep 10, 1:54 pm, ebwri..._at_gmail.com wrote:
> On Sep 10, 2:45 pm, Wiktor Moskwa <wiktorDOTmos..._at_gmail.com> wrote:
>
> > On 10.09.2007, ebwri..._at_gmail.com <ebwri..._at_gmail.com> wrote:
>
> > > I guess I'm really looking for information on what would cause a large
> > > elapsed/cpu ratio in general
> > > rather than help optimizing the specific query.
>
> > Including wait-events information in the trace file you generated
> > for the first time would be a great starting point.
> > Large elapsed/cpu ratio in general means that your session is
> > waiting for a resource instead of using it and doing its job.
>
> > --
> > Wiktor Moskwa
>
> The tkprof file has the following at the end in the summary section
>
> Elapsed times include waiting on following events:
> Event waited on Times Max. Wait Total
> Waited
> ---------------------------------------- Waited ----------
> ------------
> SQL*Net message to client 21
> 0.00 0.00
> SQL*Net message from client 21
> 0.00 0.00
>
> Is there something else in the raw trace file for which I should be
> looking?
>
> Thanks
> Eric

You should be running Statspack snapshots before and during the execution of this query, as I've stated before. Much more wait information is available to you.

David Fitzjarrell Received on Mon Sep 10 2007 - 14:31:21 CDT