Re: Native compiled code *much* slower??
- From: Dion Cho <ukja.dion@xxxxxxxxx>
- To: dreveewee@xxxxxxxxx
- Date: Sun, 27 Dec 2009 19:41:31 +0900
We are dealing with unexpected performance behavior. I don't believe that
this is easily reproducible.
I'm not sure why OP does not follow some traditional(basic) troubleshooting
procedures and post the result. For instance, wouldn't the analysis on the
difference of both cases(interpreted vs. native) give some insights?
1. 10046 trace(+tkprof)
2. PL/SQL profiling with dbms_profiler package.
3. OS system call (strace/truss/...)
4. Misc.
================================
Dion Cho - Oracle Performance Storyteller
http://dioncho.wordpress.com (english)
http://ukja.tistory.com (korean)
http://dioncho.blogspot.com (japanese)
http://ask.ex-em.com (q&a)
================================
2009/12/27 Andre van Winssen <dreveewee@xxxxxxxxx>
>
>
> Hi Steve,
>
>
>
> I ran it against 11.2 multiple times and cannot reproduce the behavior you
> describe. I see following consistent results:
>
>
>
> plsql_code_type = interpreted => 3.21 secs
>
> plsql_code_type = native => 2.9 secs
>
>
>
> (nr of iterations set to 10000)
>
>
>
> To be able to test this I had to add a missing table db_error_stack_context
> which is referenced type declarations in msc$log_p (e.g. SUBTYPE stSource IS
> db_error_stack_context.context_at%TYPE):
>
>
>
> create table db_error_stack_context
>
> (context_at varchar2(1024)
>
> ,context_value varchar2(1024)
>
> ,context_name varchar2(1024)
>
> );
>
>
>
> Regards,
>
> Andre
>
>
>
>
>
>
>
> -----Oorspronkelijk bericht-----
> Van: oracle-l-bounce@xxxxxxxxxxxxx [mailto:oracle-l-bounce@xxxxxxxxxxxxx]
> Namens Steve Baldwin
> Verzonden: vrijdag 25 december 2009 21:34
> Aan: Jared Still
> CC: oracle-l@xxxxxxxxxxxxx
> Onderwerp: Re: Native compiled code *much* slower??
>
>
>
> Here's the same but with attachments to make things (hopefully) a bit
> easier for folks to see if they see the same behaviour.
>
>
>
> Cheers,
>
>
>
> Steve
>
Other related posts:
