Re: Buffer cache statistics (ratios) and CBO SQL optimization?

Ryan Gaffuri wrote:
>
> Connor McDonald <connor_mcdonald_at_yahoo.com> wrote in message news:<3FF275DD.3F0_at_yahoo.com>...
> > Frank van Bortel wrote:
> > >
> > > Niall Litchfield wrote:
> > >
> > > > That first sentence should have read
> > > >
> > > > it just so happens that I have on my 9204 laptop a trace file that when run
> > > > through tkprof shows.... The trace file did *not* come from the laptop.
> > > >
> > > Too bad - I'd loved a laptop with a RAID-5 configuration
> > > for christmas ;-). And I'd reconfigure it for RAID 1+0
> > >
> > > --
> > > A prosperous 2004,
> > > Regards,
> > > Frank van Bortel
> >
> > Some figures from machines I've played with in the past
> >
> > Machine LIO/sec PIO/sec
> > Home PC, 500Mhz 34,950 340
> > HP, 266 Mhz 26,770 650
> > HP, 650 Mhz 74,850 2780
> > HP, 750 Mhz, 83,150 2840
> > Compaq, 700Mhz 66,470 3380
> >
> > Cheers
> > Connor
>
> Connor-- how did you calculate LIOs/second? you run a query etc... but
> how do you know how many/second? Do you just pick a long running query
> to do this?

See the hit ratio calculator on my site...Its easy enough to simply time how many consistent gets you can get per sec.

This is of course not meant to be the "definitive" logical I/O. Its just as easy to demonstrate that there is no such thing, eg, you'll find that 'select column1 from table' runs slightly faster than 'select column37 from table' due to the cost of extracting the column from the row etc etc

hth
connor Received on Wed Dec 31 2003 - 20:24:10 CST