Re: Informix vs Oracle vs DB2. SQL Query optimization.

Ian Michael Gumby wrote: > Oh BTW, a note to Daniel. Yes Daniel I read the Oracel spatial 10g > guide along with talking with several people who took Oracle's spatial > class and also used TOAD's sql explain tool. So I could see the query > plan and that's how I knew of the gaps in Oracle. > > -G > Instead of talking to "several people who took [a] class" consider talking to the instructors that teach it: You might learn something. There is a reason why documentation exists: SDO_GEOR.getRasterBlocks Returns an object of the SDO_RASTERSET collection type that identifies all blocks of a specified pyramid level that have any spatial interaction with a specified window. Source: http://download.oracle.com/docs/cd/B28359_01/appdev.111/b28398/geor_ref.htm#CHEHCBEC TOAD? Sorry but ROFLOL! Use DBMS_XPLAN or write your own based on gv$sql_plan. -- Daniel A. Morgan University of Washington damorgan@x.washington.edu (replace x with u to respond)

On Oct 11, 1:07 pm, DA Morgan <damor...@psoug.org> wrote: > Ian Michael Gumby wrote: > > Oh BTW, a note to Daniel. Yes Daniel I read the Oracel spatial 10g > > guide along with talking with several people who took Oracle's spatial > > class and also used TOAD's sql explain tool. So I could see the query > > plan and that's how I knew of the gaps in Oracle. > > > -G > > Instead of talking to "several people who took [a] class" consider > talking to the instructors that teach it: You might learn something. > Daniel, Some of the people I talked to were responsible for Oracle catching their horrible little delay in the R-Tree indexing that wasn't fixed until 10g, so the client had to tune Q-Tree indexes and limp by. And of course, the sound advice from a person who teaches an adult education course at branch college and considers himself an equal to a a tenured professor. LOL. Dude! You are so funny. BTW, while you may pick on TOAD, its a tool good enough to tell you what you need to know. Truly, you will get your biggest performance boost if you can diagnose why the query isn't using the index you thought it would. The second performance boost would be to improve your code that calls the query. The rest become a diminishing returns. (Assuming that you tuned the engine, detached the indexes, partitioned the tables, etc ...) The issue is that if you have a collection from an earlier part of the query why you can't apply the spatial index to it. I think the reason is that Oracle still has issues with derivative data types. (You call them cartridges, IDS calls them datablades.) But then again,. my client isn't your ordinary user of spatial data.