Re: Initrans
- From: "Diego Cutrone" <diegocutrone@xxxxxxxxxxxx>
- To: <Daniel.Fink@xxxxxxx>, <hamid.alavi@xxxxxxxxxxx>
- Date: Thu, 9 Sep 2004 17:31:06 -0700
You are correct Daniel, if I remember correctly, once an ITL slot has been
added to the block, the space is never reclaimed. (it's never deallocated).
That's also recorded in the 'itc' counter within the block, and can be used
to find out if the maximun number of concurrent transactions within that
block has exeeded inittrans setting.
In spite of an ITL slot only takes 24 bytes, I wouldn't change inittrans
setting for the majority of your objects. Through the years, I've only had
to change inittrans setting for a few objects (a table and some indexes).
I'd suggest you to monitor the system in search of any "enqueue TX 4 wait".
If you find any enqueue TX4 wait going on, this may suggest that may be you
have an object without enough ITL slots. (TX-4 waits may be seen due to
different reasons, one of them is due to ITL shortage, others are: multiple
sessions trying to insert the same unique key row, the first one does not
commit so the second one waits in enqueue tx4, bitmap indexes waits)
HTH
Best Regards
Diego.
----- Original Message -----
From: "Daniel W. Fink" <Daniel.Fink@xxxxxxx>
To: <hamid.alavi@xxxxxxxxxxx>
Cc: "'Oracle-L (E-mail)" <oracle-l@xxxxxxxxxxxxx>
Sent: Thursday, September 09, 2004 8:53 AM
Subject: Re: Initrans
> INITRANS is the initial number of ITEs (transaction slots) in a block.
> MAXTRANS is the maximum number of ITEs in a block (if space allows).
> Each concurrent transaction modifying a row in a block needs its own
> ITE. Unless each transaction will be modifying rows in a single block,
> setting it to the average number of tx will waste space. Small rows in
> big blocks that are modified constantly need a high number of ITEs.
> Large rows in small blocks that are rarely modified need a low number of
> ITEs. For a high concurrency object, set INITRANS to a reasonable (but
> small) number, e.g. 4. Make sure you leave enough space in the block to
> enable additional ITEs if you are wrong. I don't recall testing this,
> but I *think* that once an ITE is created, it is not removed from the
> block header.
>
> Daniel
>
> Hamid Alavi wrote:
>
>>List,
>>
>>I have heared the best number for Initras is the number of CPU but what
>>about the Freelists is there any standard for this parameter too.
>>Any idea for any standards for these two parameters? Number of CPU or
>>average number of Transaction or what?
>>Thanks,
>>
>>
>>Hamid Alavi
>>
>>Office : 818-737-0526
>>Cell phone : 818-416-5095
>>
>>
>>
>>
>>
>>======================= Confidentiality Statement =======================
>>The information contained in this message and any attachments is
>>intended only for the use of the individual or entity to which it is
>>addressed, and may contain information that is PRIVILEGED, CONFIDENTIAL
>>and exempt from disclosure under applicable law. If you have received
>>this message in error, you are prohibited from copying, distributing, or
>>using the information. Please contact the sender immediately by return
>>e-mail and delete the original message from your system.
>>===================== End Confidentiality Statement =====================
>>
>>
>>--
>>To unsubscribe - mailto:oracle-l-request@xxxxxxxxxxxxx&subject=unsubscribe
>>To search the archives - //www.freelists.org/archives/oracle-l/
>>
>>
>
> --
> To unsubscribe - mailto:oracle-l-request@xxxxxxxxxxxxx&subject=unsubscribe
> To search the archives - //www.freelists.org/archives/oracle-l/
--
To unsubscribe - mailto:oracle-l-request@xxxxxxxxxxxxx&subject=unsubscribe
To search the archives - //www.freelists.org/archives/oracle-l/
Other related posts:
