On Aug 24, 2012, at 9:31 AM, Alan Bourke wrote:
> Because they are GUIDs - globally unique. That fact is very important in
> some applications, but overkill for a lot of others.
When OpenStack was first created, it used auto-incremented keys to
identify resources such as servers, volumes, networks, etc. Later it was
redesigned to be able to scale horizontally; iow, scaling by creating separate
independent deployments that could communicate and act as a single system. Of
course, all the code to create PKs had to be changed to use UUIDs, and all the
relational code had to be updated to reference the UUID keys instead of the
integer keys. The update was a huge pain, and to this day there are both
integer and UUID keys in most tables as a result.
The worst part is that several of us argued for UUIDs from the start,
but lost that discussion to those who favored the simplicity of letting the
database handle key generation.
-- Ed Leafe
_______________________________________________
Post Messages to: ProFox@leafe.com
Subscription Maintenance: http://leafe.com/mailman/listinfo/profox
OT-free version of this list: http://leafe.com/mailman/listinfo/profoxtech
Searchable Archive: http://leafe.com/archives/search/profox
This message:
http://leafe.com/archives/byMID/profox/d88e3e2d-3e3f-4b43-8121-a953a8f8e...@leafe.com
** All postings, unless explicitly stated otherwise, are the opinions of the
author, and do not constitute legal or medical advice. This statement is added
to the messages for those lawyers who are too stupid to see the obvious.
