Notes in-line:
--
Regards
Jonathan Lewis
http://www.jlcomp.demon.co.uk
The educated person is not the person
who can answer the questions, but the
person who can question the answers -- T. Schick Jr
Next public appearance2:
March 2004 Hotsos Symposium - Keynote
March 2004 Charlotte NC - OUG Tutorial
April 2004 Iceland
One-day tutorials:
http://www.jlcomp.demon.co.uk/tutorial.html
Three-day seminar:
see http://www.jlcomp.demon.co.uk/seminar.html
____UK___February
The Co-operative Oracle Users' FAQ
http://www.jlcomp.demon.co.uk/faq/ind_faq.html
"Bjorn Augestad" <root_at_localhost.com> wrote in message
news:MkMPb.3103$Ed.49423_at_news4.e.nsc.no...
>
> Yesterday the instance, over a period of 20 hours, waited 12.000 times
> for data blocks, for a total of 1.4 million centiseconds. Large numbers
> or small, that's still a lot of waiting ;-)
>
It does sound quite large - 4 hours in 20 hours elapsed; but it may not
be so bad when averaged across the number of actual concurrent users.
I'd guess that the comments from other posters are more aimed at 'why
pick on buffer busy waits' - is this really your most significant time loss,
as it's not often something that rates highly enough that it's worth
worrying
about.
The the P3 value you quoted yesterday as the commonest occurrence was 1016,
which I think is the older version for 'lots of sessions want exclusive
locks on
the same buffer at the same time'. You've identified the tables where this
is happening,
and got some numbers for block IDs for those tables. If the numbers suggest
that
the blocking problem is at the top end of the tables and slowly moving up,
then
you need to increase the number of freelists on those tables so that you
have multiple
blocks available for insertions. (You can do this with a simple 'alter
table' in 8.1,
but I think you have to rebuild the table in 8.0).
Received on Thu Jan 22 2004 - 05:06:16 CST
