NetNews Usenet Archive 1993 #1/NN_1993_1.iso/spool/comp/sys/mac/database/1661

Text File | 1993-01-06 | 1.9 KB | 41 lines

Newsgroups: comp.sys.mac.databases
Path: sparky!uunet!stanford.edu!CSD-NewsHost.Stanford.EDU!Xenon.Stanford.EDU!michaelh
From: [e-mail] (Mike Hennahane)
Subject: Re: Looking for ideas on large mac databases
Message-ID: <[e-mail]>
Sender: [e-mail]
Organization: CS Department, Stanford University, California, USA
References: <[e-mail]>
Date:  7 Jan 93 02:39:48 GMT
Lines: 29

[e-mail] writes:
[...]
>Since we only report on historical data occasionally, I thought I would keep
>history on a tape, and copy it onto a disc when I need to use it.  I also
>thought I would keep a separate database for the most recent two years.
[...]
>But I was wondering if anyone has some thoughts on ways to manage historical
>data.  Sometimes I will want to report on time periods which overlap years
>(hence I have one big historical database instead of a few smaller ones-is it
>possible to overcome this difficulty in the new version of 4th dimension?).
>Any ideas on managing large databases on a pc would be appreciated.  Also, any
>ideas on designing a faster database would also be of interest.

if the problem is that having the extra info around is confusing, then
you might consider archiving old data to another table of the database
instead of off onto tape. that way, you can still browse the old data
in the historical section, but you only have a more manageable amount
of current data in the current section. these two tables can have
identical structures, so that you can run reports from either one, or
you can implement an un-archive function to merge some/all of the old
data back in with the current data to run reports.

as for designing a faster database, that depends mostly on the quirks
of 4d (the obvious way is not always the best), on whether you have
compiled your database, or on whether you are running the new 4d
server software.

--mike