Re: optimal size for rollback

> And your transaction waits as those checks are performed.

Your transaction waits for a lot of other things.

>
> Frankly, who cares how long it takes?

Frankly, this is what matters the most. And the answer to this question seems kinda
relevant, since we are talking about performance hit, don't you think? Remember

yourself saying "the cost of shrinkage is that performance suffers?" The good side
of that is that the space is saved. Now, savings in space may be measured in bytes,
and you tell me why not attempt to estimate the performance impact?

> It's extra work being done by the
> database (thanks for finally aknowledging it has to happen).

Please read up the thread. I never said that there was no work happening during a shrink,
You do not have to have ten years of experience to guess that there is some work.
But you said that the performance suffers, and I said it does not really suffer. Then you brought
up the horse-manure, and so on.

> And that extra
> work is entirely optional, because it doesn't happen when optimal is not
> set.

> You might consider the waits trivial. That's another of your value
> judgements, which wouldn't apply to all databases at all times. That they're
> there at all, however, is not a value judgement, but a matter of plain, hard
> fact which does apply to all databases for which optimal is in use.
>

Let me summarize what you are saying. "If you set optimal for rollback segments,
there will be something with performance as opposed to if you don't ". The message is:
"by setting the optimal you will save the space but there might be a performance impact".

Yeah, this applies to all databases, you may enjoy that. In fact, it has to, because it
is a knowledge replicated from a number of books and oracle courses. Strange of you
to assume that I or anybody else need to hear that again from you.

The next logical question is: "if we know that by setting optimal we can save N Gb of
space, how can we estimate how much the performance will suffer during extends and
shrinks?" Your answer to this is: "who cares". Interesting enough, this is where it stops
in most books and courses. Maybe because here ends theory and starts practice.

If you don't have any more valuable inputs except for the ones suggesting my professional inaptitude, let's close the thread.

Regs
AK Received on Sat Sep 21 2002 - 22:54:59 CDT