Re: 10046, Unaccounted for Time, Aix 5.3 -- How to Determine Process Pre-Emption

  • From: Cary Millsap <cary.millsap@xxxxxxxxxxxx>
  • To: elkinsl@xxxxxxxxx
  • Date: Wed, 2 Dec 2009 14:41:21 -0600

Larry,

See pages 170–173 if you have access to a copy of *Optimizing Oracle
Performance *(O'Reilly). Bottom line: unaccounted-for time in trace data on
Oracle 9.2 or beyond is almost certainly time spent preempted.

Your analysis is sweet-spot for two of the software tools we make. Check out
these if you're interested:

   - Method R Profiler (http://method-r.com/software/profiler), which
   reports on properly aggregated unaccounted-for time.
   - And *mrnl* (http://method-r.com/component/content/article/116) in the
   MR Tools suite, which allows you to drill in on exactly which specific *
   lines* of trace data are contributing the most unaccounted-for time. The
   final example on the *mrnl* web page I pointed to at the beginning of
   this paragraph is your situation on the nose.

Let me know if there's anything we can do for you...


Cary Millsap
Method R Corporation
http://method-r.com


On Wed, Dec 2, 2009 at 2:13 PM, Larry G. Elkins <elkinsl@xxxxxxxxx> wrote:

> Listers,
>
> SA's asking for assistance on how to tell if a process is being impacted
> and
> to what degree by processor pre-emption. AIX 5.3. Asked in the context of
> trouble-shooting some significant unaccounted for time issues in 10046
> trace
> files (e.g. 14 of 16 seconds unaccounted for vendor process, 300 of 700
> seconds unaccounted for vendor batch job, etc). SA's have "ruled out"
> overhead of writing the trace file so now looking at the pre-emption
> aspect.
> A bit out of my league, and I find plenty of approaches in IBM docs, and
> from people on this list (but geared more towards Solaris and other
> flavors,
> not Aix). 9.2.0.8 EE and 10.2.0.4 EE, "static" and dynamic LPAR's, seen on
> various boxes and databases.
>
> Larry G. Elkins
>
>
> --
> //www.freelists.org/webpage/oracle-l
>
>
>

Other related posts: