HP3000-L Archives

April 2003, Week 2

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Jerry Fochtman <[log in to unmask]>
Reply To:
Jerry Fochtman <[log in to unmask]>
Date:
Fri, 11 Apr 2003 13:39:02 -0500
Content-Type:
text/plain
Parts/Attachments:
text/plain (27 lines)
At 01:07 PM 4/11/2003, Ken Hirsch wrote:
>So, my question is: should we regularaly re-index our Image B-Trees?  Do
>you?  How often?

This is one of those "...it depends...".  If the dataset is fairly stable
without
a lot of add/delete, then it could be re-built less frequently that the
index for
a set that experiences frequent key add/change/delete activity.  Keep in mind
that KSAM/KSAMXL operates using a 'key' area and a 'data' area.  The key area
is dynamically maintained as a balanced tree whenever values are
added/removed.  However, the data portion is a 'first-come, first-serve'
approach so its possible that
the data for logically adjacent values are spread all over this area
causing I/O to
be extensive.  Especially when doing wild-card qualifications and it becomes
necessary for the data entries to be retrieved and evaluated against the
request.

So the answer is 'Yes", its a good idea to re-built the indexes periodically.
But at what frequency would have to be determined on an individual basis
simply because the dynamics of the data and the affect it has on data
locality are generally not easily predictable.

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

ATOM RSS1 RSS2