Re: Proposed Solution :: DB Design Dilemma!! One DB or Many DBs??

Guys, thanks for your input, but I think you're being a bit too harsh...

For what it's worth, I'm one of the most conservative DBAs you're likely to meet. I am not embarking on this new design lightly, or with reckless abandon...

Seriously, IF the GUID algorithm is solid, and always generates an absolutely unique ID, and IF I can manage the lifecycle of the rows between two databases, why couldn't a GUID be used as a PK in one database and an FK in the other?

I can think of no reason, given my assumptions hold...

Now, I just need to find a way to generate GUIDs on all DB2 version 7.2 platforms... The unique_identifer() call is interesting, but not equivalent to GUID... Does anyone know where to locate GUID generation routines that will work on Linux, AIX, OS/2, AS/400, and OS390 platforms?

Dirk and Jerry, I respect your opinions and I thank you for your warnings.
Where, exactly, do you problems with my logic, approach, or assumptions?

Many Thanks...
-Rob.

> > Rob,
> >
> > Sounds like a nightmare - both to create and maintain. Too many
> > possibilities for holes in both design and code, among other things.
>
> My thoughts exactly. I would go for one DB and one set of tables if
> possible.
>
> <mental note to self: never hire as a DBA or database developers at his shop
> ;-) >
>
> Dirk
Received on Thu Feb 14 2002 - 10:17:52 CST