Another issue with abandoning the absolute record number requirement,
besides performance, is how to accurately identify the record to be updated.
If the search item is not unique, which is often the case on Image details,
how do you make the choice?  There might be other items in the record that
could be used to distinguish a unique member of the chain, but that would
require some knowledge of the application, which would have to be
configurable in the shadowing product.  Another means would be to compare
the entire record.  Arguably, Image does not disallow duplication of even
the entire record, but in that case does it really matter if you update the
"wrong" one?  It's an interesting problem.  In either case, you would need
to read down the chain looking for the right record.  It could be pretty
costly in some databases!  You would need to consider shadowing performance
in the design of your database.

-----Original Message-----
From: F. Alfredo Rego [mailto:[log in to unmask]]
Sent: Tuesday, January 21, 2003 12:58 PM
To: [log in to unmask]
Subject: Re: ADAGER REPACK question: difference between SORTED+ and
CHAINED+


I got some private comments which I will answer on hp3000-L because
the issues are of general interest.  I will keep the identity of my
correspondent private by means of some minor editing:

>  > This is certainly a limitation of that specific shadowing product.
>
>Is there more than one on the HPe3000?

Yes.  Besides Quest's Netbase, Lund has a shadowing product, and there
are a few others that escape my mind now.


>Consider the situation of a program updating a particular record on the
>detail. Would you expect the shadow machine to execute a DBFIND and then
>traverse the chain on a DBGET?

This is a question for the shadowing folks, having to do with performance
(most likely).  "Depending on absolute entry numbers" can be one of the
prices that they choose to pay to obtain performance.  If such is the
case, then it is certainly not possible to keep the master and the shadow
repacked under different path criteria, because that would confuse the
shadowing software's dependence on absolute entry numbers.

But please keep in mind that this does not HAVE to be so.  It is a
conscious design decision to choose to depend (or not to depend) on
absolute entry numbers.  Most of the time, of course, it is appropriate
to go for performance "at any cost".  As a consequence, something else
must suffer (such is, indeed, "the cost").  Life is like that.

Eben's question is an interesting one because he wants to repack to
optimize for more than one condition (by taking advantage of the fact
that the second condition is independent of the first).  There are
many other users of AMISYS applications software and the entire
Adager team has become very interested in investigating possible
avenues that will help everybody.  Historically, we have seen a
couple of cases where a sorted repack can be more beneficial than a
chained one.  This is why I mentioned to Eben that "experimentation"
might be a good thing because "it all depends" and "your mileage may
vary"...

   _______________
  |               |
  |               |
  |            r  |  Alfredo                     [log in to unmask]
  |          e    |                           http://www.adager.com
  |        g      |  F. Alfredo Rego
  |      a        |  Manager, R & D Labs
  |    d          |  Adager Corporation
  |  A            |  Sun Valley, Idaho 83353-3000            U.S.A.
  |               |
  |_______________|

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *