| Oracle FAQ | Your Portal to the Oracle Knowledge Grid | |
 |
 | |||
| Oracle FAQ | Your Portal to the Oracle Knowledge Grid | |
|
| |||
Home -> Community -> Mailing Lists -> Oracle-L -> RE: standby config --> FAL_SERVER / FAL_CLIENT
Hi Jonathan,
Thanks for highlighting the same. I am planning to do a small testing = with maximum protection mode. I will test and share the results with the = list
Best Regards
Sriram Kumar
-----Original Message-----
From: Jonathan Lewis [mailto:jonathan_at_jlcomp.demon.co.uk]=20
Sent: Wednesday, February 11, 2004 8:49 PM
To: oracle-l_at_freelists.org
Subject: Re: standby config --> FAL_SERVER / FAL_CLIENT
This factor of 3 was a single example,=20 not a general conclusion.
The first important point that made this visible was that the sample code was committing inside a fast pl/sql loop, and contrary to the statement in the manuals about sessions waiting for lgwr, a session does not generally wait for a commit=20 (log file sync) inside a pl/sql construct.
However, if the pl/sql loop includes a distributed transaction or (it seems) has a standby database in maximum protection mode, the session suddenly DOES wait for a log file sync - and this has a big impact on performance.
Once this factor has appeared, any extra time=20 for the commit is simply down to network latency and time to write at the far end (for as many far ends as there may be) - the big issue is that you=20 moved from not waiting for commit to waiting=20 for commit.
Regards
Jonathan Lewis
http://www.jlcomp.demon.co.uk
The educated person is not the person=20 who can answer the questions, but the=20 person who can question the answers -- T. Schick Jr
Next public appearances:
March 2004 Hotsos Symposium - The Burden of Proof
March 2004 Charlotte NC OUG - CBO Tutorial
April 2004 Iceland
One-day tutorials:
http://www.jlcomp.demon.co.uk/tutorial.html
Three-day seminar:
see http://www.jlcomp.demon.co.uk/seminar.html
____UK___February
____UK___June
The Co-operative Oracle Users' FAQ
http://www.jlcomp.demon.co.uk/faq/ind_faq.html
Hi Jonathan,
Thanks for highlighting the same. I have one question on this and hope =
=3D
you would help me out. This factor of three as I understand is that if a =
=3D
normal commit takes a factor time/wait? Pls explain) of x then maximum =
=3D
protection mode would be a factor of 3x.=3D20
Also this 3x factor is for a single standby instance and would increase =
=3D
3x times for every additional setup of standby databases. Is my =3D
understanding correct?
Best Regards
Sriram Kumar
-----Original Message-----
From: Jonathan Lewis [mailto:jonathan_at_jlcomp.demon.co.uk]=3D20
Sent: Wednesday, February 11, 2004 1:45 AM
To: oracle-l_at_freelists.org
Subject: Re: standby config --> FAL_SERVER / FAL_CLIENT
Just clearing up a point on Maximum Protection and the meaning of "(does not wait till its applied)".
Isn't this the option where lgwr has to wait for confirmation that the redo has been written=3D20 into the standby log on the standby database=3D20 before writing to the online log on the primary=3D20 and acknowledging the commit.
One thing to watch out for with standby and
this option is the effect on commits within=3D20
pl/sql calls. Oracle does not wait for sync
if you issue a commit inside a pl/sql call,=3D20
issuing only one log sync write at the end=3D20
of the call - but if you switch to maximum protection,
then every commit inside the call becomes a
log sync call, and performance get much
worse. (I've had one report of a factor of
three for a piece of code doing lots of commits
inside a pl/sql loop).
Regards
Jonathan Lewis
http://www.jlcomp.demon.co.uk
The educated person is not the person=3D20 who can answer the questions, but the=3D20 person who can question the answers -- T. Schick Jr
Next public appearances:
March 2004 Hotsos Symposium - The Burden of Proof
March 2004 Charlotte NC OUG - CBO Tutorial
April 2004 Iceland
One-day tutorials:
http://www.jlcomp.demon.co.uk/tutorial.html
Three-day seminar:
see http://www.jlcomp.demon.co.uk/seminar.html
____UK___February
____UK___June
The Co-operative Oracle Users' FAQ
http://www.jlcomp.demon.co.uk/faq/ind_faq.html
Hi Prem,
Yes there is a performance overhead with maximize protection. As I said =
=3D
=3D3D
earlier, the LGWR process is going to ship the log changes to the =3D3D
standby database (does not wait till its applied) and hence It boils =
=3D3D
down to the network connectivity that exists between the primary and the =
=3D
=3D3D
standby database. Your overhead is proportional to the network =3D3D
bandwidth. Pls test it with a real load before you put in production.
Best Regards
Sriram Kumar
-- Archives are at http://www.freelists.org/archives/oracle-l/ FAQ is at http://www.freelists.org/help/fom-serve/cache/1.html ----------------------------------------------------------------- DISCLAIMER: This message contains privileged and confidential information and is =3D intended only for the individual named.If you are not the intended =3D recipient you should not disseminate,distribute,store,print, copy or =3D deliver this message.Please notify the sender immediately by e-mail if =Received on Wed Feb 11 2004 - 22:33:24 CST
=3D
you have received this e-mail by mistake and delete this e-mail from =3D your system.E-mail transmission cannot be guaranteed to be secure or =3D error-free as information could be =3D intercepted,corrupted,lost,destroyed,arrive late or incomplete or =3D contain viruses.The sender therefore does not accept liability for any =
=3D
errors or omissions in the contents of this message which arise as a =3D result of e-mail transmission. If verification is required please =3D request a hard-copy version. ---------------------------------------------------------------- Please see the official ORACLE-L FAQ: http://www.orafaq.com ---------------------------------------------------------------- To unsubscribe send email to: oracle-l-request_at_freelists.org put 'unsubscribe' in the subject line. -- Archives are at http://www.freelists.org/archives/oracle-l/ FAQ is at http://www.freelists.org/help/fom-serve/cache/1.html ----------------------------------------------------------------- ---------------------------------------------------------------- Please see the official ORACLE-L FAQ: http://www.orafaq.com ---------------------------------------------------------------- To unsubscribe send email to: oracle-l-request_at_freelists.org put 'unsubscribe' in the subject line. -- Archives are at http://www.freelists.org/archives/oracle-l/ FAQ is at http://www.freelists.org/help/fom-serve/cache/1.html ----------------------------------------------------------------- DISCLAIMER: This message contains privileged and confidential information and is = intended only for the individual named.If you are not the intended = recipient you should not disseminate,distribute,store,print, copy or = deliver this message.Please notify the sender immediately by e-mail if = you have received this e-mail by mistake and delete this e-mail from = your system.E-mail transmission cannot be guaranteed to be secure or = error-free as information could be = intercepted,corrupted,lost,destroyed,arrive late or incomplete or = contain viruses.The sender therefore does not accept liability for any = errors or omissions in the contents of this message which arise as a = result of e-mail transmission. If verification is required please = request a hard-copy version. ---------------------------------------------------------------- Please see the official ORACLE-L FAQ: http://www.orafaq.com ---------------------------------------------------------------- To unsubscribe send email to: oracle-l-request_at_freelists.org put 'unsubscribe' in the subject line. -- Archives are at http://www.freelists.org/archives/oracle-l/ FAQ is at http://www.freelists.org/help/fom-serve/cache/1.html -----------------------------------------------------------------
 |
 |