Re: Table scan and LRU caching

"Brian Peasland" <peasland_at_edcmail.cr.usgs.gov> wrote in message news:39CFBF4B.BB31840D_at_edcmail.cr.usgs.gov...
> Any block is put into the buffer cache. This does not make a difference
> if it is a full table scan or an index scan. Data is returned to the
> application from the buffer cache, so it must be read into the buffer
> cache before the application or user can read the data.

Sounds sensible...

> But the big difference lies in the location of the buffer cache list.
> Blocks read during a full table scan are placed at the Most Recently
> Used end of the list, as opposed to the Least Recently Used end of the
> list.

By doing so, I guess, the blocks of often used tables gets a better chance to "stay around" in memory, minimizing large disk reads? You'll need plenty of RAM though :-)

When do blocks get placed not at the MRU end (when first fetched)? Do blocks age off towards the LRU end unevenly?

Regards,
/fad

> HTH,
> Brian
>
>
> Dirk Mika wrote:
> >
> > Hi,
> >
> > as far as i know, blocks read during a full-table-scan are not chached.
> > But i had in mind that there is a hint to tell oracle to chache them.
> >
> > Dirk Mika
> > --
> > Bayer AG
> > WD-AED /LEV
> > D-51368 Leverkusen
> > Telefon: +49 (214) 30-50360
> > Telefax: +49 (214) 30-20031
> > Email : dirk.mika.dm1_at_bayer-ag.de
>
> --
> ========================================
> Brian Peasland
> Raytheons Systems at
> USGS EROS Data Center
> These opinions are my own and do not
> necessarily reflect the opinions of my
> company!
> ========================================
Received on Mon Sep 25 2000 - 16:59:56 CDT