Re: Explain plan execution

On 14 Apr 2006 10:50:59 -0700, "hpuxrac" <johnbhurley_at_sbcglobal.net> wrote:

>
>premmehrotra_at_hotmail.com wrote:
>> Execution Plan
>> ----------------------------------------------------------
>> 0 SELECT STATEMENT Optimizer=CHOOSE (Cost=4 Card=2 Bytes=384)
>> 1 0 SORT (ORDER BY) (Cost=4 Card=2 Bytes=384)
>> 2 1 NESTED LOOPS (Cost=3 Card=2 Bytes=384)
>> 3 2 TABLE ACCESS (BY INDEX ROWID) OF 'DM_SYSOBJECT_S' (Cos
>> t=2 Card=2 Bytes=212)
>>
>> 4 3 INDEX (RANGE SCAN) OF 'D_1F00C73D8000000F' (NON-UNIQ
>> UE) (Cost=3 Card=10)
>>
>> 5 2 TABLE ACCESS (BY INDEX ROWID) OF 'OS_DOCUMENT_S' (Cost
>> =2 Card=1 Bytes=86)
>>
>> 6 5 INDEX (UNIQUE SCAN) OF 'D_1F00C73D80000D01' (UNIQUE)
>>
>>
>> I will like to know what numbers in second "column" mean, I see there
>> are two 2's - one 3 and one 5. Does that mean step 5 (based on second
>> column) is done first, then srep3. What about two 2 steps. Which one is
>> done first?
>>
>> Is DMS_SYSOBJECT_S really the driving table for nested loop join. I
>> will be really surprised) because this table has lot more rows that
>> OS_DOCUMNET_S table. I would think OS_DOCUMENT_S should be driving
>> table.
>>
>>
>>
>> Thanks a lot,
>
>The second column is cardinality aka the number of rows.

Actually it is not and Steve's answer is correct. The cardinality is listed in card=

--
Sybrand Bakker, Senior Oracle DBA