Comments for Intermediate SQL http://intermediatesql.com Color Coded SQL, UNIX and Database Essays Wed, 22 Feb 2012 17:07:38 +0000 hourly 1 http://wordpress.org/?v=3.3.1 Comment on How ORACLE Uses Memory on AIX. Part 1: Processes by John http://intermediatesql.com/aix/how-oracle-uses-memory-on-aix-part-1-processes/comment-page-1/#comment-738 John Wed, 22 Feb 2012 17:07:38 +0000 http://intermediatesql.com/?p=22#comment-738 Hi Maxym, Nice article. Have a question of you please.... I've executed both omem.sh and ora_mem.pl against a given instance and, despite some minor differences between them, neither matches oracle sga_target setting. Is this related to what you discussed in "How AIX Paging Space works" paper? Mean, since we are not locking SGA, then AIX will grant memory to instance upon request rather than a "sga_target" size chunk upon startup. Is that so? Also, as seen below, how come requested (virtual) sometimes shows smaller than RSS? Thanks --- SQL> show parameter sga NAME TYPE VALUE ------------------------------------ ----------- ------------------------------ lock_sga boolean FALSE pre_page_sga boolean FALSE sga_max_size big integer 7008M sga_target big integer 7008M ####### ======================================================================== ===== Memory used by ORACLE instance: ======================================================================== ===== SGA: In Memory (RSS): 393 Mb Requested (Virtual): 393 Mb Segments: 28 Processes: In Memory (RSS): 693 Mb Requested (Virtual): 425 Mb Processes: 27 Total: In Memory (RSS): 1086 Mb Requested (Virtual): 818 Mb ######## Displaying memory statistics for ORACLE instance: Report mode: basic Database session information has NOT been requested ... ... ---------- ---------- ---------- ---------- ---------- ----- -------------------- PID Total InMem Paging Virtual Pages COMMAND ---------- ---------- ---------- ---------- ---------- ----- -------------------- TOTAL: 1377.68 1317.92 59.76 1358.47 s M Processes: 48 Hi Maxym,
Nice article. Have a question of you please…. I’ve executed both omem.sh and ora_mem.pl against a given instance and, despite some minor differences between them, neither matches oracle sga_target setting.
Is this related to what you discussed in “How AIX Paging Space works” paper? Mean, since we are not locking SGA, then AIX will grant memory to instance upon request rather than a “sga_target” size chunk upon startup. Is that so? Also, as seen below, how come requested (virtual) sometimes shows smaller than RSS? Thanks

—
SQL> show parameter sga

NAME TYPE VALUE
———————————— ———–
——————————
lock_sga boolean FALSE
pre_page_sga boolean FALSE
sga_max_size big integer 7008M
sga_target big integer 7008M

#######

======================================================================
=====
Memory used by ORACLE instance:
========================================================================
===

SGA:
In Memory (RSS): 393 Mb
Requested (Virtual): 393 Mb
Segments: 28

Processes:
In Memory (RSS): 693 Mb
Requested (Virtual): 425 Mb
Processes: 27

Total:
In Memory (RSS): 1086 Mb
Requested (Virtual): 818 Mb

########

Displaying memory statistics for ORACLE instance: Report mode: basic
Database session information has NOT been requested
…
…

———- ———- ———- ———- ———- —–
——————–
PID Total InMem Paging Virtual Pages
COMMAND
———- ———- ———- ———- ———- —–
——————–
TOTAL: 1377.68 1317.92 59.76 1358.47 s M Processes:
48

]]> Comment on What is the difference between SQL Profile and SPM Baseline ? by Maxym Kharchenko http://intermediatesql.com/oracle/what-is-the-difference-between-sql-profile-and-spm-baseline/comment-page-1/#comment-731 Maxym Kharchenko Wed, 25 Jan 2012 21:50:28 +0000 http://intermediatesql.com/?p=523#comment-731 Thanks, Kerry. That makes perfect sense now :-) Thanks, Kerry.

That makes perfect sense now :-)

]]> Comment on What is the difference between SQL Profile and SPM Baseline ? by Kerry Osborne http://intermediatesql.com/oracle/what-is-the-difference-between-sql-profile-and-spm-baseline/comment-page-1/#comment-730 Kerry Osborne Wed, 25 Jan 2012 21:38:29 +0000 http://intermediatesql.com/?p=523#comment-730 Hi Maxym, Just stumbled across this post again and noticed you had asked me a question. Sorry I didn't notice before now. I have seen Baselines be disregarded, even without such extreme conditions as a specified index having been removed. The reason for this is that Baselines attempt to apply enough hints to limit the choices the optimizer has to a single plan, but there are situations where the set of hints is not sufficient to actually force the desired plan. What I mean is that the hints will eliminate virtually all possibility but there still may be a few that are valid and so it's possible to get a different plan. In fact, I have even seen situations where the act of creating a Baseline causes the plan to change. This is clearly not intended behavior but it can happen. I think I blogged about that once. At any rate, in these cases where the optimizer arrives at a different plan than was expected, the optimizer uses a plan generated completely without the hints (or possible picks another accepted Baseline and applies it's hints). This behavior is very different from SQL Profiles which blindly apply the hints. I should also note that SQL Profiles can also be created manually with the same set of hints that would be used in a Baseline and there is a MOS note regarding how to do this. I have blogged about that as well here: http://kerryosborne.oracle-guy.com/2010/07/sqlt-coe_xfr_sql_profilesql/ Hi Maxym,

Just stumbled across this post again and noticed you had asked me a question. Sorry I didn’t notice before now.

I have seen Baselines be disregarded, even without such extreme conditions as a specified index having been removed. The reason for this is that Baselines attempt to apply enough hints to limit the choices the optimizer has to a single plan, but there are situations where the set of hints is not sufficient to actually force the desired plan. What I mean is that the hints will eliminate virtually all possibility but there still may be a few that are valid and so it’s possible to get a different plan. In fact, I have even seen situations where the act of creating a Baseline causes the plan to change. This is clearly not intended behavior but it can happen. I think I blogged about that once. At any rate, in these cases where the optimizer arrives at a different plan than was expected, the optimizer uses a plan generated completely without the hints (or possible picks another accepted Baseline and applies it’s hints). This behavior is very different from SQL Profiles which blindly apply the hints. I should also note that SQL Profiles can also be created manually with the same set of hints that would be used in a Baseline and there is a MOS note regarding how to do this. I have blogged about that as well here:

http://kerryosborne.oracle-guy.com/2010/07/sqlt-coe_xfr_sql_profilesql/

]]> Comment on What are SQL Profiles and why do we need them ? by Maxym Kharchenko http://intermediatesql.com/oracle/what-are-sql-profiles-and-why-do-we-need-them/comment-page-1/#comment-723 Maxym Kharchenko Thu, 05 Jan 2012 02:20:26 +0000 http://intermediatesql.com/?p=496#comment-723 John, thanks for your kind words John, thanks for your kind words

]]> Comment on What are SQL Profiles and why do we need them ? by John http://intermediatesql.com/oracle/what-are-sql-profiles-and-why-do-we-need-them/comment-page-1/#comment-722 John Wed, 04 Jan 2012 09:29:01 +0000 http://intermediatesql.com/?p=496#comment-722 Good article clearly written - also good point re long term support Good article clearly written – also good point re long term support

]]> Comment on What are SQL Profiles and why do we need them ? by Michael Fontana http://intermediatesql.com/oracle/what-are-sql-profiles-and-why-do-we-need-them/comment-page-1/#comment-718 Michael Fontana Thu, 22 Dec 2011 18:37:40 +0000 http://intermediatesql.com/?p=496#comment-718 Good article, by the way. Good article, by the way.

]]> Comment on What are SQL Profiles and why do we need them ? by Michael Fontana http://intermediatesql.com/oracle/what-are-sql-profiles-and-why-do-we-need-them/comment-page-1/#comment-717 Michael Fontana Thu, 22 Dec 2011 18:28:33 +0000 http://intermediatesql.com/?p=496#comment-717 While you allude to the potential maintenance downside to profiles, you forget to mention how they could become a permanent crutch. Rather than fix the application or design, you'll now be doing a standardized task of maintaining would could turn into hundreds of specialized instructions which will have to be supported release to release, platform to platform. Also, their behavior COULD change due to environmental or software changes by the vendors. While you allude to the potential maintenance downside to profiles, you forget to mention how they could become a permanent crutch. Rather than fix the application or design, you’ll now be doing a standardized task of maintaining would could turn into hundreds of specialized instructions which will have to be supported release to release, platform to platform. Also, their behavior COULD change due to environmental or software changes by the vendors.

]]> Comment on ora_cpu.pl tool for Linux by Maxym Kharchenko http://intermediatesql.com/linux/ora_cpu-pl-tool-for-linux/comment-page-1/#comment-498 Maxym Kharchenko Tue, 27 Sep 2011 16:19:41 +0000 http://intermediatesql.com/?p=608#comment-498 Thanks, JC Thanks, JC

]]> Comment on ora_cpu.pl tool for Linux by Jean-Christophe DAUCHY http://intermediatesql.com/linux/ora_cpu-pl-tool-for-linux/comment-page-1/#comment-497 Jean-Christophe DAUCHY Tue, 27 Sep 2011 14:37:39 +0000 http://intermediatesql.com/?p=608#comment-497 Thanks awesome, it was already really nice for AIX.. since there is not much tool for AIX (except nmon of course). Thanks awesome, it was already really nice for AIX.. since there is not much tool for AIX (except nmon of course).

]]> Comment on About by Vishith Kumar http://intermediatesql.com/about/comment-page-1/#comment-487 Vishith Kumar Fri, 23 Sep 2011 13:26:16 +0000 http:/?page_id=2#comment-487 Max, Your Style of presentation and explantion of oracle concepts is awesome man.Keep posting more in the blog Regards, Vishith Max, Your Style of presentation and explantion of oracle concepts is awesome man.Keep posting more in the blog

Regards,
Vishith

]]>