Niall Litchfield wrote:
> "Ryan Gaffuri" <rgaffuri_at_cox.net> wrote in message
> news:1efdad5b.0302060916.354b850d_at_posting.google.com...
>
>>So I take it anytime I see something by Niemic, I should take it with
>>a grain of salt. Alot of people hate his performance tuning book and
>>now that I have read several others I see the weaknesses.
>
>
> That is putting words into my mouth that aren't quite there. Anyone can make
> errors, and people can say things ineffectively. See my comment about
> execute immediate in another thread for examples of this :( . As it happens
> I don't think that the article is terrible, but I do think it is unclear and
> in places not apparently well thought out. I'd hate people to think that it
> is about personality, its about clear , substantiated best practice. To that
> end take 'anything' you read by anyone with that pinch of salt. this is a
> pretty good approach.
>
> I'll give you another example
>
> There are two well known names, whom I believe take contradictory views on
> the desirability of Automatic Segment Space Management in 9i. Who is right -
> I don't know, but it ought to be possible to set up valid tests in your
> environment to see if ASSM gives better or worse results than setting your
> own freelists and freelist groups. What would undoubtedly be a bad approach
> would be to say ' Ah but <insert recognised name here> says <insert oracle
> resource here> that ASSM is universally bad/good'.
Excellent advice, Niall as ever.
But Niemic's stuff is still almost always bad. :-)
Regards
HJR
Received on Fri Feb 07 2003 - 04:26:20 CST
