Re: latch-free SCN scheme (10.1.0.3)

From: Connor McDonald <hamcdc_at_yahoo.co.uk>
Date: Tue, 15 Feb 2005 23:58:20 +0000 (GMT)
Message-ID: <20050215235820.79503.qmail@web86904.mail.ukl.yahoo.com>

There appears to be two enhancements..

In 9i, they introduced "log_parallelism" which is more than a single redo allocation latch. In 10g, this parameter becomes hidden because the amount of parallelism is auto-decided/tuned by Oracle. So there's 3 parameters:

_log_parallelism
_log_parallelism_dynamic
_log_parallelism_max

Also in 10g, there a new thing called private redo. All pretty much conjecture on my part here but it looks like some of the log buffer gets created in such a way that sessions can directly copy into it without requiring latches. At a guess, "10 private strands" means up to 10 sessions can get benefit from this mechanism at any given instance...from x$ksppi

_log_private_mul             - Private strand multiplier for log space preallocation
_log_private_parallelism     - Number of private log buffer strands for zero-copy redo
_log_private_parallelism_mul - Active sessions multiplier to deduce number of private strands

I'm sure Tanel or Steve may have greater insight.

Cheers
Connor

Mladen Gogala <gogala_at_sbcglobal.net> wrote:

> In my alert.log I have the following text:
>
> Mon Feb 14 20:29:29 2005
> Starting ORACLE instance (normal)
> LICENSE_MAX_SESSION =3D 0
> LICENSE_SESSIONS_WARNING =3D 0
> Picked latch-free SCN scheme 2
> KCCDEBUG_LEVEL =3D 0
> Using LOG_ARCHIVE_DEST_10 parameter default value as USE_DB_RECOVERY_FILE_D=
> EST
> Autotune of undo retention is turned on.
> Dynamic strands is set to TRUE
> Running with 1 shared and 10 private strand(s). Zero-copy redo is FALSE
>
>
>
> What is "latch-free SCN scheme"? As far as I am aware, any transaction that=
> needed
> to increase SCN, needed to acquire latch that was protecting SCN. Is it sti=
> ll the=20
> case? V$LATCHNAME, of course, tells a different story:
>
> 1 select name,latch# from v$latchname
> 2 where name like '%SCN%'
> 3* order by 1
> SQL> /
>
> NAME LATCH#
> ------------------------- ----------
> batching SCNs 110
> change tracking consisten 155
> t SCN
>
> change tracking optimizat 154
> ion SCN
>
> flashback SCN barrier 159
> flashback hint SCN barrie 161
> r
>
>
> NAME LATCH#
> ------------------------- ----------
> lgwr LWN SCN 106
> mostly latch-free SCN 105
> ping redo on-disk SCN 108
> redo on-disk SCN 107
>
> 9 rows selected.
>
>
> What's the deal? If SCN acquisition is really latch free, ie does not requi=
> re=20
> previous latch acquisition, that would be a great performance enhancement.=20
> If that is so, is that true for RAC as well? What about local and global SC=
> N?
> My world is falling apart! Moreover, what is "zero-copy redo"?
> Once upon a time, there was a parameter which was determining the maximum
> size of redo entry that was written directly to log buffer. Everything
> greater then that was first formatted in the process buffer, space was
> then allocated in the redo buffer (by acquiring redo allocation latch, of
> course), and when redo copy latch was acquired, the user buffer was copied
> into the log buffer. Am I correct in my understanding that all processes=20
> need to acquire only allocation latch and that they will write directly
> into the log buffer? No copy latch necessary? Jonathan, please help!
> --=20
> Mladen Gogala
> Oracle DBA
>
>
> --
> http://www.freelists.org/webpage/oracle-l
>

Connor McDonald
Co-author: "Mastering Oracle PL/SQL - Practical Solutions" ISBN: 1590592174 web: http://www.oracledba.co.uk
web: http://www.oaktable.net
email: connor_mcdonald_at_yahoo.com

Coming Soon! "Oracle Insight - Tales of the OakTable"

"GIVE a man a fish and he will eat for a day. But TEACH him how to fish, and...he will sit in a boat and drink beer all day"

Do you Yahoo!?
The all-new My Yahoo! - What will yours do? http://my.yahoo.com

--
http://www.freelists.org/webpage/oracle-l

Received on Tue Feb 15 2005 - 19:01:12 CST