[Foundation-l] [Internal-l] Relocation Announcement

Mon Sep 24 19:48:24 UTC 2007

On 9/24/07, Mark Bergsma <mark at wikimedia.org> wrote:
> Anthony wrote:
> >>>> And that's just one HTTP request. It would be way worse if we sent all
> >>>> the very many DB/memcached/DNS/etc lookups necessary to build a single
> >>>> wiki page over the atlantic...
> >>> If you didn't pipeline them.  But why in the world would you do that?
> >> Because of the "update everywhere problem". Data of which multiple
> >> copies exist either needs all copies updated atomically (impractical due
> >> to the latency we were talking about) or needs to synced/merged later,
> >> with all the fun conflict resolution problems you will get.
> >>
> > I can't figure out what your response has to do with the question.
> > What is it we're talking about, again?
>
> Because the information needs to live in one place. The entire world is
> editing it. How to decide where to put it? Making multiple copies is not
> the answer.
>
Making multiple copies of what, the database?  Yeah, as I understand
it what you guys now have is way better.  Multiple read-only caches,
which can sometimes be out of date by a few seconds, and one central
read-write db with the "official" copy of everything.

This discussion does bring up the point that the centralized db should
probably be in a better location, though.  Tampa doesn't seem to have
very good peering with the rest of the world, or even with other parts
of Tampa!