Note in-line.
--
Regards
Jonathan Lewis
http://www.jlcomp.demon.co.uk
The Co-operative Oracle Users' FAQ
http://www.jlcomp.demon.co.uk/faq/ind_faq.html
April 2004 Iceland http://www.index.is/oracleday.php
June 2004 UK - Optimising Oracle Seminar
"Douglas Hawthorne" <douglashawthorne_at_yahoo.com.au> wrote in message
news:%f5ac.130893$Wa.95465_at_news-server.bigpond.net.au...
> However, this appears to contradict the blocksizes given in Appendix E
where
> the blocksizes are given as 2K except for the following segment names:
> (1) ORDRCLUSTER_QUEUE 16K
> (2) ROLL_SEG 8K (most likely their name for the automatic UNDO segments).
>
> If Appendix E is more accurate than digging around in the scripts of
> Appendix B, then a blocksize of 2K was chosen for some reason in nearly
all
> the cases.
>
I haven't read the paper - but if a lot of the tables use hashing instead
of indexing, then a smaller block size could reduce the CPU cost of
hash-collisions.
Received on Tue Mar 30 2004 - 02:15:08 CST
