Hi Milan,

I agree with most of your comments, but the point that I was trying to make is that the classical stuff scales quite a long ways. Stack overflow is ranked by Alexa somewhere at 1,600. Popular second tier web shops (not Amazon, but eg. Fnac in France or Bol in the Netherlands) rank similar or lower. So there are perhaps 1,000 web properties that really need a scale out solution, the rest can make do with running eg. Firebird in lots of memory.

Assume that the scale out solution is sold for $25K initial license plus four years of maintenance for 20% x 25K (this is the original SQLServer price point), the web db market is only $9K x 1,000 web properties, equates $9 mln. IBM and Oracle price a $100K initial license, so that would make the web db market ~$40 mln. That is not enough to get VC's to invest the $30 mln that is needed these days to build a new serious software company.

There are a few ways to overcome this problem, just to name a few:
- convince number 1,001 to 10,000 on the Alexa list that they need a scale out database right from the start because they may become a top-1000 player
- find uses for scale out databases other than web properties
- don't sell a cloud database but a web solution, including all the surrounding engineering in $ 1 mln projects
- ...

Paul

--- In Firebird-Architect@yahoogroups.com, Milan Babuskov <milanb@...> wrote:
>
> paulruizendaal wrote:
> > Hi bird hats, here's some interesting reads:
> >
> > There's a write-up about stackoverflow.com architecture here:
> > http://highscalability.com/blog/2009/8/5/stack-overflow-architecture.html
> >
> > It is different in the sense that it uses an all-Microsoft stack and that the database is more scale-up than scale-out (SQLServer running on 8 cores in 48GB of memory).
>
> Although I admire stackoverflow as a website and concept, and I use it a
> lot, I listened to Jeff Atwood on many podcast(s) and even read a few
> posts on his blog, and I see that architecture as quite amateur (not to
> say any harsh words). He's thinking on the same level you would expect
> from some grad student without any field experience. His partner, Joel
> Spolsky is quite different story, but I'm afraid he doesn't have much
> say in technical "department", especially since they haven't hit the
> brick wall yet. IMHO, of course. Sooner or later they will hit the limit
> of such system and will not be able to scale it anymore, and then
> they're going to be in for a "big rewrite". OTOH, they do have a niche
> market, so maybe they will not reach this limit ever.
>
> Just my $.02 - I hope I don't get my SO account canceled over this :)
>
> --
> Milan Babuskov
>
> ==================================
> The easiest way to import XML, CSV
> and textual files into Firebird:
> http://www.guacosoft.com/xmlwizard
> ==================================
>