Re: Suggestions for first exam ...

"Anurag Varma" <avdbi_at_hotmail.com> wrote in message news:H5Qdb.3038$6y3.1848_at_news01.roc.ny...
>
> "Jonathan Lewis" <jonathan_at_jlcomp.demon.co.uk> wrote in message
 news:bl61qb$mec$1$8302bc10_at_news.demon.co.uk...
> --snip--
> > "Richard Foote" <richard.foote_at_bigpond.com> wrote in message
> > news:rGgdb.126227$bo1.111770_at_news-server.bigpond.net.au...
> > >
> > > Coincidently again (first the last time), I've noticed a new article
 by
> > > Jonathan Lewis at dbazine which nicely describes why the unbalanced
 Oracle
> > > index is myth of utter crappy proportions (perhaps my complaint had
 an
> > > effect ?)
> > >
> >
> > I wrote it because I came across yet another bloody
> > "expert" explaining that sequences cause Oracle's
> > indexes to get extremely deep on the right so that
> > indexes need to be rebuilt.
> >
> > There is another article coming out soon (I think)
> > that points out why it is usually a waste of time
> > to rebuild indexes at all - and gives an example
> > showing how a regular rebuild can make things
> > worse.
>
>
> The article by Mike Hordila is published in oracle own site:
> http://www.oracle.com/oramag/webcolumns/2001/index.html?auto_index.html
>
> Now how does Oracle itself let this guy publish (in their own website) a
 totally untrue article showing the ;) fascinating
> diagram of what unbalanced indexes look like.
>
> :)
> In fact you should see his site http://members.rogers.com/mhordila/
> he's got loads of useless and myth creating articles out there.
>

Hi Anurag

Oh boy, this all get's worse.

I visited www.dbazine.com today to show someone the quality of some of the articles (they didn't believe half the horror stories I was telling them). Anyway, a popup screen appeared suggesting I enrol and download a web-book on Oracle indexes by the world's leading experts. This should be interesting I thought and proceeded to download the thing. Although some of the articles were quite good (our very own Jonathan Lewis Bitmap article for a start), some were the appalling articles I mentioned earlier. Mike Hordila's horror story was there, as was John Weeg's debacle (recommend that indexes have 4 extents and then rebuild to clean up that wasteful deleted stuff) and one by Don Burleson who describes how doubling the index block halves the consistent reads (true but did it actually run faster, get it actually generate less O/S I/O calls, I don't want to get into the discussion of appropriate block sizes, especially for file based sub-systems but larger index blocks isn't necessarily the way ...)

There are of course two unfortunate issues here. By "hiding" (if that's the word) appalling articles among decent ones, it kinda gives them a level of acceptance and legitimacy that's dangerous and a trap for the unwary. It also suggests that the writers of the decent articles agree and support the views and technical merit of those articles that surround them when that might not necessarily be the case.

I'm not sure what the solution is but to raise these issues as best one can and hope that at least some take notice...

I've said no in the past but boy, writing a book oneself certainly has some merits.

Cheers

Richard Received on Mon Sep 29 2003 - 06:10:21 CDT