Re: FTS, LRU/MRU and X$BH

I believe that blocks are inserted midpoint in the buffer cache and not at the LRU end. This started in 8i (I think) and the touchcount (TCH) determines which end the block migrates to. That does not explain why you are not seeing *some* blocks but I thought we should be aware of this change in behavior.

(On crackberry in a train so don't have access to check)

John

On 1/12/09, Martin Berger <martin.a.berger_at_gmail.com> wrote:
> Rich,
>
> you can also check V$SEGMENT_STATISTICS
> http://download.oracle.com/docs/cd/B14117_01/server.101/b10755/dynviews_2060.htm#REFRN30220
> what kind of reads where done on this.
>
>
> hth
> Martin
>
>
>
>> Ahhh...yes, that would account for this. I also have a non-parallel
>> version
>> of the query that I'll check to verify as well.
>>
>> Thanks, Jaromir!
>>
>> Rich
>>
>>> Hi Rich,
>>>
>>>
>>>> I confirmed that the docs at
>>>> http://tinyurl.com/a6n4bz say FTSing will cause blocks to be put
>>>> on the
>>>> LRU end of the buffer cache, so I expected to at least see them at
>>>> some
>>>> status.
>>>
>>> Please check the DEGREE of the table. One explanation could be you
>>> use a
>>> parallel select and therefore a direct path read.
>>> http://download.oracle.com/docs/cd/B19306_01/em.102/b25986/oracle_database.htm#sthref1922
>>> In doubts, re-run the select with NO_PARALLEL hint.
>>>
>>
>> --
>> http://www.freelists.org/webpage/oracle-l
>>
>>
>
>

-- 
Sent from my mobile device

John Kanagaraj <><
http://www.linkedin.com/in/johnkanagaraj
http://jkanagaraj.wordpress.com (Sorry - not an Oracle blog!)
** The opinions and facts contained in this message are entirely mine
and do not reflect those of my employer or customers **
--
http://www.freelists.org/webpage/oracle-l