Subject: | |
From: | |
Reply To: | F. Alfredo Rego |
Date: | Wed, 14 May 1997 07:04:03 -0600 |
Content-Type: | text/plain |
Parts/Attachments: |
|
|
Jerry Fochtman <[log in to unmask]> wrote:
>CIU will help retain entry chronology for unaffected search keys.
>Furthermore, the performance analysis which Wirt Atmar and I did, and which
>I presented at the New Orleans conference, showed that CIU can provide
>upwards to 800% better performance in changing a single search field value
>on a 16-key dataset. If any of the key fields which are not being changed
>have sort fields, using CIU saves even more overhead as there is no sort
>field search to re-insert the entry like there would be if DBDELETE/DBPUT
>are used.
I can attest to the thorough work that Wirt and Jerry did while testing CIU
and I endorse Jerry's recommendations.
In fact, a lot of Wirt's and Jerry's final testing was done in Sun Valley,
using Adager's laboratory setup back in the 1991/1992 time frame. I have
fond memories of evening walks under the crisp mountain skies while
discussing the tremendous performance possibilities of CIU.
Jerry was one of the earliest supporters of the concept and he was brave
enough to be on my side as I was beating on Doug Spreng's head at the
Washington Interex conference in 1985 (Doug Spreng was the head of HP's CSY
at the time; no pun intended). Wirt then took the lead with a series of
superb technical articles on the CIU subject. The rest is history.
It appears that "marketing" of the CIU concept has not reached as wide an
audience as it deserves. How about some articles on the subject for the
next SigIMAGE Newsletter (hint, hint)? As the editor, I am standing by...
_______________
| |
| |
| r | Alfredo [log in to unmask]
| e | http://www.adager.com
| g | F. Alfredo Rego Tel 208 726-9100
| a | Manager, R & D Labs Fax 208 726-2822
| d | Adager Corporation
| A | Sun Valley, Idaho 83353-3000 U.S.A.
| |
|_______________|
.
|
|
|