Re: Oracle's Myth: keep tables and indexes in separate tablespaces

Howard J. Rogers wrote:

>Mark Townsend wrote:
>
>
>
>>Howard J. Rogers wrote:
>>
>>
>>>Jack wrote:
>>>
>>>
>>>
>>>
>>>>Look:
>>>>http://otn.oracle.com/deploy/performance/pdf/opt_storage_conf.pdf
>>>>
>>>>You might have a look at Juan Loaiza's article "Optimal Storage
>>>>Configuration Made Easy" http://technet.oracle.com/deploy/performance/
>>>>for a
>>>>different approach to the problem. Since this is on technet, it appears
>>>>to be Oracle's latest recommendation.
>>>>
>>>>
>>>Sorry Jack: you're wrong on this, though I suppose it depends on what you
>>>mean by 'recommendation'.
>>>
>>>
>>I don't believe that Jack is wrong.
>>
>>
>
>It's not a matter of belief. If the course notes say it in black and white,
>then he's wrong.
>
>
>
>>In fact, if the courses ARE teaching
>>it differently,
>>
>>
>
>No, they've been teaching it in exactly the same way since at least version
>8.0.
>
>
>
>>then I think the course needs to be urgently updated -
>>can anybody send me a reference number, and I'll follow up on it.
>>
>>
>
>They just *have* been updated. Check out 9i Performance Tuning. They've just
>released version 2.1 of that document, and it contains the 'separate tables
>from indexes' "recommendation" in Chapter 3. I taught it just last week,
>and my notes were version 2.0, and the students had version 2.1... that's
>how recent the update of the material was... I didn't actually get told a
>new revision had been issued. Fortunately, the changes were minor (and this
>wasn't one of them).
>
>I can't give you the course code, I'm afraid, because I haven't actually
>cleared my desk out yet, so the manuals are still sitting in the office.
>
>But I can promise you that Fundamentals I is still "recommending" the same
>thing, too.
>
>I only know it because every time I teach it I have to spend 15 minutes
>explaining why the course notes are wrong, usually to a bunch of
>initially-sceptical faces. And it's been that way for years, so I doubt
>anything will change in a hurry.
>
>If you care to wait until early next week, I can have the precise course
>codes and page numbers for you (unless Richard wants to jump in
>beforehand).
>
>
>
>>Note that ASM (Automated Storage Management) in Oracle Database 10g
>>implements the SAME methodology as the preferred (and only supported)
>>best practice.
>>
>>
>
>Oh dear. SAME isn't a magic bullet, either. Frying pan and fire time, I
>suppose. Still it's better than nothing.
>
>Regards
>HJR
>
>

After I got home from teaching last night I read Mark's posting and thought about responding but the Sandman was beating me with a large bag of sand so I went straight to bed.

But, a bit belatedly, a quick comment for Mark.

Oracle Corp. would do itself a huge favor if it gave far more attention to the content of what it teaches in its classes, what it puts its name to on its various web sites, on the advice given out in its name by Oracle Press, and by some person's associated with IOUG.

This is not to say that all is horrible and rotten. For the most part it is remarkably good. But there are some things that seem to persist year after year, version after version, that are just not true. A quick review of postings here on usenet with key words like "Myth" and "Mythology" will point you to some of the more blatantly obvious of them.

-- 
Daniel Morgan
http://www.outreach.washington.edu/ext/certificates/oad/oad_crs.asp
http://www.outreach.washington.edu/ext/certificates/aoa/aoa_crs.asp
damorgan_at_x.washington.edu
(replace 'x' with a 'u' to reply)