I believe you're right - I was thinking
about the 'cache fusion' bit where
the BSP is proclaimed to serve
consistent blocks - but still bounces
a block off the disc if it gets too
many requests for a CR block in
too short a period of time.
--
Jonathan Lewis
http://www.jlcomp.demon.co.uk
Host to The Co-Operative Oracle Users' FAQ
http://www.jlcomp.demon.co.uk/faq/ind_faq.html
Author of:
Practical Oracle 8i: Building Efficient Databases
Screen saver or Life saver: http://www.ud.com
Use spare CPU to assist in cancer research.
Howard J. Rogers wrote in message <3b968754_at_news.iprimus.com.au>...
>
>"Jonathan Lewis" <jonathan_at_jlcomp.demon.co.uk> wrote in message
>news:999702126.17719.0.nnrp-01.9e984b29_at_news.demon.co.uk...
>>
>> You may have far fewer problems with
>> pinging - if you define pinging as writing
>> a buffer block to disc because a different
>> instance wants to take it in exclusive mode.
>>
>
>If you define pinging like that, then there should be no pinging
whatsoever,
>since in RAC blocks are transferred across the interconnect, not via disk.
>
>It's true I may be missing a scenario or three, but I've yet to find one
>that requires a write to disk before the requesting Instance can obtain the
>resource.
>
Received on Wed Sep 05 2001 - 15:21:29 CDT
