Home » RDBMS Server » Performance Tuning » Big database migration
Big database migration [message #559619] Wed, 04 July 2012 23:36 Go to next message
majstoru
Messages: 68
Registered: October 2006
Location: Serbia
Member
Hi,

I have one big database which I need to migrate to Oracle, because it rocks with big databases, instead of other databases and when I was made transfer software and all works great except one more think. During this process I found that Oracle normally fill log & undo table, and my question is how to migrate (or can I migrate) database to oracle without filling undo database (deactivate this process) and after that to put database to work normally, because I just need to transfer data as is and from that point Oracle goes on...

Thanks in advanced...
Re: Big database migration [message #559620 is a reply to message #559619] Wed, 04 July 2012 23:52 Go to previous messageGo to next message
BlackSwan
Messages: 26766
Registered: January 2009
Location: SoCal
Senior Member
Please read and follow the forum guidelines, to enable us to help you:

http://www.orafaq.com/forum/t/88153/0/

You can not turn off redo & undo.
Either accept it or use different RDBMS.
Re: Big database migration [message #559623 is a reply to message #559620] Thu, 05 July 2012 00:46 Go to previous messageGo to next message
Michel Cadot
Messages: 68625
Registered: March 2007
Location: Nanterre, France, http://...
Senior Member
Account Moderator
... and for instance, you could migrate using transportable tablespaces.

Regards
Michel
Re: Big database migration [message #560514 is a reply to message #559619] Fri, 13 July 2012 14:54 Go to previous messageGo to next message
alan.kendall@nfl.com
Messages: 163
Registered: June 2012
Location: Culver City, California
Senior Member
Some Ideas for inproving performance.
1) You can turn archive log mode off (requires a shutdown/startup) or delete archivelogs as needed during the load from within rman. I find that this did NOT improve performance.
2) A large part of time spent within loading tables is the redo generated from indexes (not the tables) so consider dropping indexes on the largest tables and then create the indexes after the tables are loaded.
3) Use sqlloader to load the largest tables since it is so fast.
4) Increase the pga_aggregate_target at the system level to decrease disk sorts when creating indexes.
ECSCDAQ > alter system set sga_target=20g scope=spfile sid='*';

System altered.

or at the session level
ECSCDAQ > alter session set workarea_size_policy='MANUAL';

Session altered.

ECSCDAQ > alter session set sort_area_size=1000000000;

Session altered.

ECSCDAQ > alter session set hash_area_size=1000000000;

Session altered.

ECSCDAQ > alter session set sort_area_retained_size=1000000000;

Session altered.
Re: Big database migration [message #560518 is a reply to message #560514] Fri, 13 July 2012 15:45 Go to previous message
majstoru
Messages: 68
Registered: October 2006
Location: Serbia
Member
Thanks to all repliers for help...
Previous Topic: The percentage of time spent by sessions waiting on redo operations
Next Topic: Response very slow
Goto Forum:
  


Current Time: Fri Mar 29 04:01:20 CDT 2024