different because the first query loaded the rows into memory.
John
[EMAIL PROTECTED] wrote:
All the columns in the table are not null ...
here is an interesting test I ran ...
several Direct_Path_Write and Read waits with large elasped times
are showing up in the 10046 trace data on this query due to the
full table scan.
Just deleted statistics on this and fwiw ...
With Statistics ....
=====================
SELECT Statement Optimizer=Choose (Cost=225 ...)
SORT (Order By) (Cost=225 ...)
Table Access (Full) of 'table1'
Physical Reads = 433
Without Statistics ..
======================
SELECT Statement Optimizer=Choose
Table Access (By Index Row) of 'table1
Index ( Full Scan ) of 'concatenated index' (Non-Unique)
Physical Reads = 0
Mike
-----Original Message-----
Sent: Thursday, January 23, 2003 6:29 PM
To: Multiple recipients of list ORACLE-L
....
Anyone have any problems with the CBO not using
a index when you know it is faster by forcing a hint ?
I have set the following ...
Solaris Oracle Version 8.1.7.4
block size = 8
DB_FILE_MULTIBLOCK_READ_COUNT = 8
mode = Choose
also using Tim Gormans 90 and 50 values for the other optimizer parms.
Select col1, col2, col3, blah1, blah2 from table order by col1, col2, col3;
Concatenate index on col1, col2, col3.
-- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Ora NT DBA INET: [EMAIL PROTECTED] Fat City Network Services -- 858-538-5051 http://www.fatcity.com San Diego, California -- Mailing list and web hosting services --------------------------------------------------------------------- To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
