Re: Optimization

From: Holger Baer <holger.baer_at_science-computing.de>
Date: Fri, 14 Jan 2005 18:06:29 +0100
Message-ID: <cs8u6l$bn5$1@news.BelWue.DE>

Peter Mubaiter wrote:
> Niall Litchfield schrieb:
>

>> I disagree. It *might* be nothing to worry about, but it *might* - and
>> more probably does - mean one of two things

>
>
> <snip>
>
> I think that most of the database out there will have a high value for
> wait event "SQL*Net message from client". I know nearly no application
> which opens a session, requests data, fetches the data and closes the
> session again (that'll keep this specific wait event low).
>

Webbased applications (e.g jsp or asp) do exactly this. If you don't have a fully grown Application Server, that's what you get.

 > But of course, there can be a network which is too slow, but the
 > statement "I have got a high value of the wait event" is too less
 > information...
 >

I agree on that. It's too little information to tell if there is a problem at all. But there might well be - so one shouldn't discard that wait as insignificant. If this was from a batch job it would be significant.

My 2 eurocent (at the time of writing much more than then US ones) ;-)

Holger Received on Fri Jan 14 2005 - 11:06:29 CST