PS_COMM_SCTN_DATA & ccltrgen.sqr

[skyway <skyway@USWEST.NET>]

This particular problem involves the SQR that creates the
letter extracts within the PeopleSoft Higher
Education module. The ccltrgen.sqr populates several tables,
however, the PS_COMM_SCTN_DATA
seems to be most impacted as the setting in the process
scheduler Go - Peopletools - Process Scheduler -
Use - Process Definitions
would indicate [ Y 2000 10 6 :STANDRD_LTR_TBL.ADMIN_FUNCTION
]  The "Y" setting  produces
approximately 52 rows of data per record during the processing
and storage of the viewable fields after
processing a ccltrgen.sqr , these 52 rows per record include
only the name and address of the recipient of
the letter. The table PS_COMM_SCTN_DATA increases quite
quickly in size, which can be of a great
surprise. The size of this table can grow from 6 megs to 350
in a few days.

The other element, is the [ Y 2000 10 6
:STANDRD_LTR_TBL.ADMIN_FUNCTION]  if the "Y" is
changed to "N" then the size of the storage increases at least
by 50% approximately.

Question?  Considering that this one table could require 4-8
gigs all by itself,  per year for ccltrgen.sqr
viewable records, is any other institution having experiences
like this. The answer to the situation could be
handled in many ways. However, wondering if the efficiency of
the whole process isn't a bit wasteful taking
up all of this disk space.
As there are considerable fields which hold white space.
Perhaps the big question is how important is it to
have viewable history of history of letters and fields used to
generate letters, if the business need is high, then
disk space must be allocated. This type of table growth is
quite significant and cannot be overlooked during
planning and allocation of system resources. Thus the interest
in bringing this into the public eye. Any work
around, suggestions, solutions greatly appreciated.






begin:vcard
n:Oltmanns;Larry
x-mozilla-html:TRUE
org:
adr:
version:2.1
email;internet:skyway@mn.uswest.net
title:Consultant
fn:Larry  Oltmanns
end:vcard