RE: *Measuring sql performance (elapsed time and scalability) by number of logical reads

In the trace data context, those things listed below (hash value calculation, memory management, traversing hash chains, etc.) will show up as time in the c field. ...Not as time in a "wait" event.

I don't think the categorization "service" vs. "wait" helps people's thinking very much. A better way to think of the time being spent from Oracle's perspective is "kernel code path" vs. "OS call code path". Roughly speaking, the c field on a dbcall maps to kernel code path execution, and the ela field on so-called "wait" lines maps to OS call code path execution.

Notice that an OS call duration includes both waiting and service time. This is the reason I don't like to use the word "wait interface".

Cary Millsap
Hotsos Enterprises, Ltd.
http://www.hotsos.com
Nullius in verba  

Hotsos Symposium 2007 / March 4-8 / Dallas Visit www.hotsos.com for curriculum and schedule details...

-----Original Message-----
From: oracle-l-bounce_at_freelists.org
[mailto:oracle-l-bounce_at_freelists.org] On Behalf Of Radoulov, Dimitre Sent: Tuesday, May 02, 2006 4:15 PM
To: Andrey Kriushin
Cc: oracle-l_at_freelists.org
Subject: Re: *Measuring sql performance (elapsed time and scalability) by number of logical reads

[...]
> The second is IFFS with 610 LIO in 72513 us.
> Total is 1523 LIO..... AND 223005 us. Not 83118 us, right? Looks like
> we've found ~140 msec. And those 140 msec are not just hash value
> calculation, but also memory management (chunk allocation) in PGA,
> probably growing PGA itself etc.
>
> Also, as somebody already mentioned, LIO time can vary significantly.
> 10605/913 ~= 11.6 us and 72513/610 ~= 119 us. The length of hash
chains
> may count, for example. Well, on quiet instance that might be row
> processing time inside the block as well and not only the time to pin
the
> buffer.

[...]

Thank you for your time and valuable feedback. Anyway, correct me if I'm wrong, but I would expect those to be considered
wait events, not service time.

Regards,
Dimitre

--
http://www.freelists.org/webpage/oracle-l


--
http://www.freelists.org/webpage/oracle-l