On Mon, Apr 30, 2012 at 07:55:00PM +0100, Simon Riggs wrote:
> On Mon, Apr 30, 2012 at 6:38 PM, Bruce Momjian <bruce@momjian.us> wrote:
>
> > I would love to see a layout of exactly where these things make sense,
> > similar to what we do at the bottom of our documentation for "High
> > Availability, Load Balancing, and Replication":
> >
> > http://www.postgresql.org/docs/9.1/static/different-replication-solutions.html
> >
> > Users and developers just can't seem to get the calculus of where things
> > make sense into their heads, me included.
> >
> > For example, you said that "MM replication alone is not a solution for
> > large data or the general case". Why is that? Is the goal of your work
> > really to do logical replciation, which allows for major version
> > upgrades? Is that the defining feature?
>
> Good question.
>
> The use case, its breadth and utility are always the first place I
> start. I'm in the middle of writing a presentation that explains this
> from first principles and will be discussing that at the PgCon
> meeting. It's taken a long time to articulate that rather than make
> leaps of assumption and belief.
Yep, it is the "assumption and belief" that always confuses me.
-- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB
http://enterprisedb.com
+ It's impossible for everything to be true. +