> I think you have made the point by detailing your approach: it leads to
> complicated code in several places.
>
> To me, the hallmark of a great approach is that complex stuff can be
> epxressed in little code. The unix v6 kernel was less than 10,000 sloc
> and the code reads almost like a book most of the time.

Ok, then Spread is for your - JGroups is definitely bigger. :)

>> True, but in this case you do not need to implement a lot, just take
>> Sequoia and run your cluster. You have dynamic addition of nodes
>> there too - a database adapter is required to implement
>> backup/restore service.
>
> I thought sequioa was a write-everywhere solution too.

They support write everywhere as well as partitioning, writes are
redirected to the nodes which have data.

> How does it deal
> with 1,000+ updates/second and staying consistent at the same time?

Yes, writing my email I had that question as well. However, since curren
production Sequoia version is 2.something and they have thrown away the
3.x tree, I'm going to wait for 4.x and Tungsten project to check that.

Roman