RE: Why "Separating Data and Indexes improves performance" is a myth?

From: Jacques Kilchoer <Jacques.Kilchoer_at_quest.com>
Date: Wed, 28 Apr 2004 10:18:56 -0700
Message-ID: <B5C5F99D765BB744B54FFDF35F60262109F877DA@irvmbxw02>

So it is possible in recent versions! I just assumed it wasn't because = of the allowed values for DB_nK_cache_size for creating a tablespace = with a different blocksize.

One thing that would definitely give you an error in 8.1 was that if you = have a locally managed autoallocate tablespace you were unable to create = any object in it.

-----Original Message-----
From: Niall Litchfield=20

Jonathan wrote
> I believe Stephan Haisley has built a 9i database
> with a 7.5K block size (not for production reasons,
> of course).

SYS 27-APR-04_at_block>show parameter block

NAME                                 TYPE        VALUE
------------------------------------ ----------- =
----------------------------
db_block_buffers                     integer     0
db_block_checking                    boolean     FALSE
db_block_checksum                    boolean     TRUE
db_block_size                        integer     5120
db_file_multiblock_read_count        integer     16

SYS 27-APR-04_at_block>show release
release 1001000200
SYS 27-APR-04_at_block>

5 and 10 are such round numbers I thought I might choose the highest = common divisor for my block size...

if anyone has suggestions as to how to achieve interesting errors with = this setup the db will be around for a day or so.

---

Please see the official ORACLE-L FAQ: http://www.orafaq.com

---

To unsubscribe send email to: oracle-l-request_at_freelists.org put 'unsubscribe' in the subject line.

--
Archives are at http://www.freelists.org/archives/oracle-l/
FAQ is at http://www.freelists.org/help/fom-serve/cache/1.html
-----------------------------------------------------------------

Received on Wed Apr 28 2004 - 12:15:43 CDT