"Jan Wieck" <JanWieck@Yahoo.com> wrote in message
news:3E942A99.A6102F2D@Yahoo.com...
> Ron Peacetree wrote:
> > [...]
> > The lack of two phase =commit= is a also a potential performance
hit
> > in comparison to DB products that have it, but the more important
> > issue there is that there are SMP/distributed apps that really
can't
> > work acceptably unless a DB product has two phase commit.
> >
> > The three "biggies" in DB land, SQL Server, Oracle, and DB2, have
both
> > features. I suspect that PostgreSQL will need to as well...
>
> Ron, do you actually have some ideas how to do 2 phase commits?
> Especially things like how to re-lock during startup after a crash
and
> the like? Or is your knowledge in reality only buzzwords collected
from
> high glossy tradeshow flyers?
>
If "some ideas" means "do I know how to code it into PostgreSQL right
now", the answer is no. If "some ideas" means "do I understand the
general problem at a technical level well enough to be thinking about
the algorithms and datastructures needed to support the functionality"
the answer is yes.
So I'd say a fair response to your questions is that my knowledge is
in between the two extremes you've described, but probably closer to
the first than the second ;-). We can have a private email discussion
on the topic if you wish.
