My question was specific to RAC: With a larger block size, is one not
increasing the chances of pinging for the blocks? Or I dont need to worry
about it with Cache fusion? A new worry for me now is that, though the
block size was changed from 2K to 8K, the recommendation was to keep the
SGA the same?
Anyways, I am still setting up the database, and I plan to create a few
tables in tablespaces of varied block sizes, and test it myself. During
stress testing, I expect to see a lot of waits.
A new question: All recommendations that I have read so far is to make the
datablock block size equal to the file system block size? But what if my
datafiles are raw disks?
Thanks
Raj
Charlie_Mengler@Hom
eDepot.com To: Multiple recipients of list
ORACLE-L <[EMAIL PROTECTED]>
Sent by: cc:
[EMAIL PROTECTED] Subject: Re: 2k and 8k block size
February 14, 2003
11:09 AM
Please respond to
ORACLE-L
There was a presentation at the recent HOTSOS conference which showed
a performance gain by placing indexes in tablespaces with a "LARGE"
blocksize.
HTH & YMMV
HAND!
Michael Fontana
<[EMAIL PROTECTED] To: Multiple
recipients of list ORACLE-L <[EMAIL PROTECTED]>
et> cc:
Sent by: Subject: Re: 2k and 8k
block size
[EMAIL PROTECTED]
02/10/2003 01:59
PM
Please respond to
ORACLE-L
Reading my oracle 9i new features , there is now something called:
Multiple Database Block Size Support. In other words, it is no longer
necessary
to commit to a single block size for your database. I'd suggest looking
into this, as
it may now be possible to test using different blocksizes for different
tablespaces.
At 01:04 PM 2/10/2003 -0800, [EMAIL PROTECTED] wrote:
An 8.0.6 ops database being migrated to 9i RAC database on Solaris 8.
The
8.0 database had an block size of 2k. Multiblock read count is 8.
This is
an OLTP database, all transactions almost always work on a single
row. None
of the row sizes exceed the block size. Hardly any chained or
migrated
rows. Now, during this migration, it was recommended to change the
block
size to 8k. And my concerns are:
1. Would this not increase pinging across the instances? With 9i and
interconnect and cache fusion, this should be reduced considerably,
so, is
it a mute point?
2. This could alter the execution plans of some statements. Would
some,
maybe a very few of the statements not prefer a full table scan?
Any other positive or negative effects of this increase in block
size?
Thanks
One Paranoid DBA
--
Please see the official ORACLE-L FAQ: http://www.orafaq.net
--
Author:
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).
