Robert Hodges wrote:
Hi everyone,
Some of the previous data type conversations reminded me of a question
that's been on my mind for a while. What do you think of dropping
auto-increment keys and replacing them completely with UUIDs?
Auto-incrementing is one of the banes of horizontal scaling so it seems
reasonable to substitute something that works across partitions and data
copies. Or at least to give application developers an opportunity to
suppress it.
Thanks, Robert
--
Robert Hodges, CTO, Continuent, Inc.
http://scale-out-blog.blogspot.com
Actually, I'd kind of like Auto-Increment to be something I could use
aside from just on a Key
For you Postgres guys out there you already know what I want, Sequences
But for Keys, yes, I'd very much like a UUID key, especially if it would
be guaranteed unique (which if I remember right mysql had some issues
with a while back), even if the same data went into two servers at the
exact same time
Not so sure as a replacement for incrementing keys, but looking at a
distributed system, ie multi-master, yeah, very useful.
So I'm really seeing this as more of a request for UUID as an
self-filling data type than for replacing auto-increment, yeah?
Would cover both bases, single master web applications can still have
their old familiar auto-increment, and the new cloud guys can get unique
identifiers that can scale?
_______________________________________________
Mailing list: https://launchpad.net/~drizzle-discuss
Post to : [email protected]
Unsubscribe : https://launchpad.net/~drizzle-discuss
More help : https://help.launchpad.net/ListHelp
