Re: logical file access paths

The access path is a tree structure of sorts. The pointers to the individual
records are at the bottom (leaf nodes) of the tree. When you change the key
value of a record it removes the existing leaf node, inserts a new node in
the appropriate place and then for both affected places potentially
rebalances the branch above it until it reaches the point where it does not
need to rebalance. The indexing technology that IBM uses is call a Binary
Radix index for the indexes that you create in DDS. Note: There can be more
than two nodes attached to any given node at a time. Eventually if you add
or delete enough things will get rebalanced and you do not have to worry
about degenerative tries...

Here is a link which contrasts EVI indexes vs. Binary Radix Indexes.
http://www.as400.ibm.com/developer/bi/evi.html

If you are really really interested then there is a wonderful text written
by Donald Knuth. "The Art of Computer Programming/Sorting and Searching."
Right now the publisher has sold out (new edition which is being sucked up
by all the geeks in the world :-) The whole three volume set is available
though and I personally can not recommend it enough!!!
http://www.amazon.com/exec/obidos/ASIN/0201485419/qid=955632533/sr=1-1/103-1
197673-3691861

The three books in this series will broaden any computer programmer's
horizons and as Bill Gates said, "If anyone can read through these books
then they should send me their resume." The is a lot of incredible
information and these books are the most compact representation of distilled
computer programming wisdom I have read.

I hope this helps
Eric


______________________________________________
Eric N. Wilson
President
Doulos Software & Computer Services
2913 N Alder St.
Tacoma WA 98407


----- Original Message -----
From: "Colin Thorpe" <Colin@IBS-PUBLIC-SERVICES.CO.UK>
To: <RPG400-L@midrange.com>
Sent: Thursday, April 13, 2000 1:13 AM
Subject: logical file access paths


> Thanks to all who responded to the original post....
>
> >The initial build has to read the entire file, build the key table,
(Access
> >Path), then sort the Access Path and write it to a file.  This initial
read
> >of the entire 10mil record file takes a while.  Adding/Deleting/Updating
> >record keys is done rather quickly on a record by record basis.  Don't
you
> >just love the power of the AS400s native data base.
>
>
> A couple of points for clarity........
>
> 1. This is not a real issue, it is something we are discussing in our
office
> at the moment.
>
> 2. I understand the above but,
>     If you change/update a record that is part of the key (ie order/access
> path etc) why does it not have to
>     rebuild the order/access path to accomodate this new record order and
> therefore take some time.
>
> Thanks
>
>
> Colin
>
> +---
> | This is the RPG/400 Mailing List!
> | To submit a new message, send your mail to RPG400-L@midrange.com.
> | To subscribe to this list send email to RPG400-L-SUB@midrange.com.
> | To unsubscribe from this list send email to RPG400-L-UNSUB@midrange.com.
> | Questions should be directed to the list owner/operator:
david@midrange.com
> +---


+---
| This is the RPG/400 Mailing List!
| To submit a new message, send your mail to RPG400-L@midrange.com.
| To subscribe to this list send email to RPG400-L-SUB@midrange.com.
| To unsubscribe from this list send email to RPG400-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---