Re: SGA Free memory

  • From: Christo Kutrovsky <kutrovsky.oracle@xxxxxxxxx>
  • To: jonathan@xxxxxxxxxxxxxxxxxx
  • Date: Mon, 13 Feb 2006 14:20:04 -0500

On a related note, have you guys noticed the very large memories on a
"round" address with KSMCHCOM = 'permanent memor' and KSMCHCLS =
'perm' ?

SQL> select * from sys.x_$ksmsp where KSMCHCOM = 'permanent memor' and
KSMCHCLS = 'perm' and ksmchsiz > 10000000;

ADDR                   INDX    INST_ID   KSMCHIDX   KSMCHDUR KSMCHCOM 
       KSMCHPTR           KSMCHSIZ KSMCHCLS   KSMCHTYP KSMCHPAR
---------------- ---------- ---------- ---------- ----------
---------------- ---------------- ---------- -------- ----------
----------------
0000002A97C66E18         13          1          1          1 permanent
memor  00000001BC0CD000   15899128 perm              0 00
0000002A97C669F8         25          1          1          1 permanent
memor  00000001BB0CD000   15596208 perm              0 00
0000002A97C66790         32          1          1          1 permanent
memor  00000001BA0CD000   15933896 perm              0 00
0000002A97C66370         44          1          1          1 permanent
memor  00000001B90CD000   15933896 perm              0 00
0000002A97C67388         90          1          1          1 permanent
memor  00000001B80CD000   13792800 perm              0 00
0000002A97C68A28        117          1          1          1 permanent
memor  00000001B70CD000   15902000 perm              0 00
0000002A97C68870        122          1          1          1 permanent
memor  00000001B60CD000   15851552 perm              0 00
0000002A97C68660        128          1          1          1 permanent
memor  00000001B50CD000   15109080 perm              0 00
0000002A97C68030        146          1          1          1 permanent
memor  00000001B40CD000   15886048 perm              0 00
0000002A97C692B0        185          1          1          1 permanent
memor  00000001B30CD000   14680096 perm              0 00
0000002A97BF2C78       6093          1          1          1 permanent
memor  00000001AF0CD000   15937376 perm              0 00

This looks to me like some kind of reserved memory, as I remember
seeing a couple of these disappear in favor of other areas.

Jonathan what other articles would you recommend for shared pool insides ?

--
Christo Kutrovsky
Senior Database/System Administrator
The Pythian Group



On 2/11/06, Jonathan Lewis <jonathan@xxxxxxxxxxxxxxxxxx> wrote:
>
> Thanks for the reference - interesting paper.
>
> Regards
>
> Jonathan Lewis
> http://www.oracle.com/technology/community/oracle_ace/ace1.html#lewis
>
> The Co-operative Oracle Users' FAQ
> http://www.jlcomp.demon.co.uk/faq/ind_faq.html
>
> Cost Based Oracle: Fundamentals
> http://www.jlcomp.demon.co.uk/cbo_book/ind_book.html
>
>
> ----- Original Message -----
> From: "Luca Canali" <Luca.Canali@xxxxxxx>
> To: <oracle-l@xxxxxxxxxxxxx>
> Sent: Friday, February 10, 2006 8:15 AM
> Subject: RE: SGA Free memory
>
>
> Hi Jonathan,
>
> >    loop
> >        detach some items from end of library cache LRU
> >        distribute items to shared pool free list - coalescing where
> possible
> >        check for large enough chunk on free list
> >        return if allocation succeeds
> >    end loop
> >
> >    There may be a limit on the number of times round the loop -
>
> Apparently the limit is 5 iterations (see Russell Green, OOW Sep. 2005)
>
> Regards,
> Luca
>
> -----------
> Luca Canali
> Information Technology Department
> European Organization for Nuclear Research
> CERN
> CH-1211 Geneva 23
> --
> //www.freelists.org/webpage/oracle-l
>
>
>
>
> --
> //www.freelists.org/webpage/oracle-l
>
>
>


--
Christo Kutrovsky
Senior Database/System Administrator
The Pythian Group
--
//www.freelists.org/webpage/oracle-l


Other related posts: