Re: SQL> select class,count from v$waitstat; show data block contention

Normally a recycle buffer cache is used for objects that tend to over take the buffer cache in normal use (like rapid inserts into an object, e.g. history type objects). So an recycle pool is kept small to leave enough buffers available for the other objects. If for some reason we need to access a buffer again that was flushed from the recycle pool, multiple processes could want to read the same block causing an bbw. That is the only scenario that I can imagine ....

Anjo.

Nuno Souto wrote:

> On Tue, 13 Mar 2001 09:47:43 GMT, Anjo Kolk <k.kolk_at_chello.nl> wrote:
>
> >
> >
> >From my experience, waits on data blocks means either that you are missing
> >freelists (many processes are inserting in to the same object) or the
> >buffer cache is too small too keep your current working set in and you have
> >to read them from disk all the time.
> >
>
> Thanks for the info, Anjo. Would having blocks in buffer pool RECYCLE
> cause a similar problem (or make it more likely to happen)?
>
> Cheers
> Nuno Souto
> nsouto_at_bigpond.net.au.nospam
> http://www.users.bigpond.net.au/the_Den/index.html
Received on Tue Mar 13 2001 - 08:41:37 CST