Re: Is there a performance limit of extents?

Following that point up though,
what's the largest number of tablespaces anyone has ever set up, and what were the side effects.

I can claim 896 - with 3,500 files - and on a parallel server that was a real killer in 8.0.3.

Problems with excessive (segment) checkpointing and SMON interference etc....

Has anyone gone to real extremes with 8.1 ?

--

Jonathan Lewis
Yet another Oracle-related web site:  http://www.jlcomp.demon.co.uk

Mark D Powell wrote in message <8dl5q4$cvt$1_at_nnrp1.deja.com>...


>In article <956170241.12883.2.nnrp-03.9e984b29_at_news.demon.co.uk>,

>  "Jonathan Lewis" <jonathan_at_jlcomp.demon.co.uk> wrote:

>>

>>

>> If you have large numbers of relatively small

>> extents which are NOT adjacent to each other

>> then the Query co-ordinator can end up passing

>> out a very large number of very small chunks to

>> query slaves, which can result in a silly amount

>> of overhead.  It is therefore sensible to aim for

>> a relatively low number of extents per segment

>> on objects likely to be targeted by Parallel Queries.

>>

>>

>Excellent point! But since PQ works best with larger objects this is

>another reason to separate very large objects into separate tablespaces

>from very small objects to begin with. I am a strong believer in trying

>to standardize the extent sizes as it does make DBA life easier.

>--