Re: technical reasons for splitting data-model into schemes

From: Dave O'Keeffe <DOKeeffe_at_n0spam.cableinet.co.uk>
Date: Tue, 21 Apr 1998 21:27:37 +0100
Message-ID: <353D0139.3635@n0spam.cableinet.co.uk>

Michael Pelzelmeyer wrote:
>
> Are there any technical reason - besides better organization for
> developers, dba's - for splitting a big data-model into schemas?
> The only technical consequence I know is the default tablespace and
> temporary tablespace.
> Does somebody know any splitting criterias? (e.g. group by business
> area, ...)
> TIA
> Michael

Michael,

Do you mean splitting it by schemas or by tablespaces ? There are a number of reasons to split the tables implementing the data model into different tablespaces (disk i/o balancing, tablespace backups, data/index splitting for simpler backup/recovery, read only tablespaces to name but a few).
To split the tables into different schemas (equivalent of different "owners" of the tables) might make sense if they are used by different applications (a bit like a distributed database) but I can't see it myself.

Dave O'Keeffe
--
For email replys remove "n0spam." from the above address Received on Tue Apr 21 1998 - 15:27:37 CDT