RE: Re[2]: Physics of the FILTER operation within SQL_PLANE.

Edgar,

>One thing that bothers me, is don't you think that in future version
>of Oracle, it should provide finer granularity of 'cpu service' event?

Exactly!

>Wouldn't it be nice, if we'd have cpu service wait events similar to
>'db file sequential read', but for cpu service, like
>'sorting'/'merging',
>now we can see this kind of information in v$session_longops, but
>unfortunately this information is not logged in trace files.

I believe you *could* work out some of this information from Statistics in v$sesstat. I am using a kinda snapshot of v$sessstat to look at further details while diagnosing performance issues in addition to 10046 trace and snapshots of v$session_event/v$session_wait. And yes - it would be a good idea to log this in the trace file. For example, if only Oracle Corp would enhance the trace generator to log _differences_ in v$sesstat at strategic intetervals when say the trace level is set to 16 (i.e. (4+8)+4). This would be in addition to the normal information present in 10046.

I believe that there may additional OS related information present when you turn on TIMED_OS_STATISTICS on supported platforms/oracle versions. I haven't really seen the effect of this on Production systems yet. The gurus of course would be able to add to this!

Hth,
John Kanagaraj <><
DB Soft Inc
Phone: 408-970-7002 (W)

http://tahiti.oracle.com - Manuals for DBAs http://www.biblegateway.com - Manual for Life

• The opinions and facts contained in this message are entirely mine and do not reflect those of my employer or customers **

  ---

  Please see the official ORACLE-L FAQ: http://www.orafaq.com

  ---

  To unsubscribe send email to: oracle-l-request_at_freelists.org put 'unsubscribe' in the subject line. -- Archives are at http://www.freelists.org/archives/oracle-l/ FAQ is at http://www.freelists.org/help/fom-serve/cache/1.html

  ---