Re: question about automatic undo management

From: Jonathan Lewis <jonathan_at_jlcomp.demon.co.uk>
Date: Thu, 23 Jan 2003 17:39:45 -0000
Message-ID: <b0p9vv$mri$1$8302bc10@news.demon.co.uk>

It comes back to classifying the systems you are familiar with.

90% of the systems that I see are massively overpowered for the work they get done - but who cares, because the hardware is relatively cheap and the job gets done quickly enough to satisfy the user. But there are still a few systems where high-precision implementation matters. (And where disks are not cheap because of the cost of the extra back-plane, cabinet, fibre controller, floor re-inforcement and air-con).

--
Regards

Jonathan Lewis
http://www.jlcomp.demon.co.uk

Coming soon a new one-day tutorial:
Cost Based Optimisation
(see http://www.jlcomp.demon.co.uk/tutorial.html )

Next Seminar dates:
(see http://www.jlcomp.demon.co.uk/seminar.html )

____England______January 21/23
____USA_(CA, TX)_August


The Co-operative Oracle Users' FAQ
http://www.jlcomp.demon.co.uk/faq/ind_faq.html





Alex Filonov wrote in message
<336da121.0301230901.71133f4_at_posting.google.com>...


>Jonathan,

>

>I was asking this question specifically, because I don't understand

>why reducind I/O is so important. It was 7 years ago, I remember

>that. Now, with fast dist systems and plenty of memory it's not

>that important. My experience of 3 last years (Oracle Apps on

>a very powerful multi-CPU machines): no case of I/O contention.

>Lots of cases of network contention, latch contention and, oh

>yeah, ORA-1555. I'd trade those for a little of I/O contention...

>

>Best regards,

>

>Alex.

>

Received on Thu Jan 23 2003 - 11:39:45 CST