From: | Daniel Farina <daniel(at)heroku(dot)com> |
---|---|
To: | Josh Berkus <josh(at)agliodbs(dot)com> |
Cc: | pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: Changing the concept of a DATABASE |
Date: | 2012-05-22 20:15:45 |
Message-ID: | CAAZKuFaa-fW_Nb6DA8QqKHF7yYbo_8y1KMZs0xH3ngcDRhsTuw@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Tue, May 22, 2012 at 10:56 AM, Josh Berkus <josh(at)agliodbs(dot)com> wrote:
> I'm not arguing that we don't have users who would like interdatabase
> queries, especially when they port applications from MySQL or MSSQL. We
> have a lot of such users. However, we *also* have a lot of users who
> would like to treat separate databases as virtual private instances of
> Postgres, and there's no way to satisfy both goals. We have to choose
> one route or the other.
I think the idea that a physical machine where catalogs are physically
(shared-everything) co-located is one that will not stand for long as
part of a useful contract between a user and the database. I'd really
like to avoid an extra tier of functionality that exists only for
databases that happen to land on the same physical machine.
I think any inter-database feature should work identically between two
databases across a network as two machines on one machine/cluster.
Transparent optimizations to deal with the special case of physical
co-location are not contrary to that contract, but I don't have a
sense of how important those optimizations would be before getting a
lot of the usability issues figured out. Right now, it seems to me
that getting interdatabase usability feeling better is already pretty
hard.
--
fdr
From | Date | Subject | |
---|---|---|---|
Next Message | Daniel Farina | 2012-05-22 20:24:39 | Re: Schema version management |
Previous Message | Vik Reykja | 2012-05-22 19:58:11 | Re: Add primary key/unique constraint using prefix columns of an index |