Home » SQL & PL/SQL » SQL & PL/SQL » join operations in warehouse environment
join operations in warehouse environment [message #20230] Wed, 08 May 2002 06:22 Go to next message
bechir
Messages: 23
Registered: November 2001
Junior Member
what join operation (merge join, nested loops, or hash join) fits warehouse environment the most?
Re: join operations in warehouse environment [message #20284 is a reply to message #20230] Fri, 10 May 2002 23:19 Go to previous messageGo to next message
Sreedhar
Messages: 30
Registered: March 2000
Member
hi bechir...

DW queries are purely dependent on the condition..
in DW you alomost avoide indexes and lookup tables.. all you have is.. summary tables.. and tables with Huge and Huge data...

what ever the query you write you need to explain plan and optimize the query... whcih ever way it does fit... so no one can tell which to use when... it is you need to decide...

thanks
-Sreedhar
Re: join operations in warehouse environment [message #20333 is a reply to message #20230] Tue, 14 May 2002 06:40 Go to previous message
bechir
Messages: 23
Registered: November 2001
Junior Member
Sreedhar,
Here is my specific concern:
In case where join tables include indexes, using the RULE hint will enforce Oracle optimizer to use nested loops. These loops return the first matching rows quickly to users-they don't wait for the whole set of records to be selected. Therefore, they are more suited for online users and less efficient for warehouse environment. I'm saying this because, as I understand, warehouse users need the whole set of records and not incremental return of rows.
your opinion?
thanks!
Previous Topic: SQL*Loader
Next Topic: Selecting multiple columns in Pl/SQL
Goto Forum:
  


Current Time: Wed Apr 17 20:45:31 CDT 2024