Re: Segment management auto clause

"Howard J. Rogers" <howardjr2000_at_yahoo.com.au> wrote in message news:tBEwa.34614$1s1.503185_at_newsfeeds.bigpond.com...
>
>
> Get off the case, Richard! You like ASSM? Be my guest.

Oh come on Howard, don't be like that. I'm only playing with ya !!

It's not that I "like" ASSM, it's just that I "like" to give them a chance.

Cost / Benefit Analysis and if they win (sometimes) great and if they lose (othertimes) then too bad.

>
> What does RAC have to do with large tables? You're likely to have large
> tables in a RAC. That's all. No smart-arse comments needed. Of course you
> can have large tables *not* in a RAC, and I didn't suggest otherwise.
>
> But if a RAC is characterised by large tables, then it stands to reason
that
> ASSM might be a go-er, because the % overhead is not going to be
> significant.

Sorry, read it again but I still don't get it. Big tables - RAC, big tables RAC ... nup ?

>
> But all of that is beside the point anyway, because with my shares in
> Seagate, I really don't give two hoots about 3% extra disk space
> utilisation. It's full table scan times and
kiss-goodbye-to-the-buffer-cache
> that I dislike.

You cruel bugger, don't you know ASSM are listening :)

It depends, it depends, it depends ...

But all things being equal (and that means one hell of a lot to the point of being meaningless), yes ASSM does cost ya with FTS. Again, the key question, is the "cost" affordable and does it outweigh any possible advantages.

But what if things aren't equal ?

And what makes something equal or not equal ?

And what the hell am I going on about ..... ?

>
> >
> > We have some very large tables at my current site and ummmm, no RAC.
I'm
> > sure that there are many, many, many other sites with large tables that
> are
> > RACless.
> >
> > Like most ..
> >
> > I wouldn't call your statement a sweeping generalisation, but an
incorrect
> > one :)
> >
>
> Yes, well you would, wouldn't you. Never let the words I write get in the
> way of a good argument, huh?

What argument ?

Love ya :)

Richard Received on Thu May 15 2003 - 06:44:28 CDT