Hi Patrick:  Thanks for the note.  I will do as you suggest. 

To Mr. Rego and many others:  Thank you for your input.  Based on
Patrick's analysis (which in turn was based on the HOWMESSY report I
sent him) and in light of the actual benefits to be gained, I will
forego the SORTED+ repack, and opt instead for the CHAINED+ repack.  

-----Original Message-----
From: Patrick Mullen [mailto:[log in to unmask]] 
Sent: Tuesday, January 21, 2003 1:49 PM
To: Eben Yong
Subject: RE: [HP3000-L] ADAGER REPACK question: difference between
SORTED+ and CHAINED+

Eben,

This will wrap but if you stretch the window to full screen width it
should
unwrap:

SERVICE            Det12905753 6432488  49.8%  ( 7767910)   7S!CLAIM#
147    2.09    2.32    1.04    1.40  19.4%   1.35
                                                               SERV#
1    1.00    0.00    1.00    1.00   0.0%   1.00
                                                             S MEMBER#
24551   62.14  175.15    9.35   44.24  69.6%   4.73

MEMBER# is the only path that would truly benefit from a repack, anyway.

I would repack member# with the chained option.  the claim# should not
be adversely affected by that but check it with another howmessy
afterwards.

I based this determination on the elongation value (in this case).
An elongation of 1 is perfect but I doubt that you could improve
much on 1.35, so repack MEMBER# and I'll bet that CLAIM# won't
suffer.

By the way, I would allocate 6-12 hours for this repack.  The
locality (elongation=4.73) is horrible, hence the long estimate.

It will probably be closer to 6 hours but I have to guess-timate
12 so you don't get caught unprepared, timewise.

Patrick

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