So this page is work in progress. It may take me a little while to get this done.
So I will start this page with a description of the distributed service bus that I work on. It uses a Super Peer architecture to organize its topology. It is hierarchical and uses that hierarchy to route messages, this is called Reactive Routing. So an enterprise or network of nodes will look like a Christmas tree. There as the concept of message direction as well, up, down, and all. So messages can be made to move up or down the hierarchy. Services and consumers on this network have logical addresses. Messages can be addressed to a specific entity or broadcast to many entities.
That is an example of what I am sure is just one implementation of a Distributed Service Bus. Let�s see if we can't break that down into some defining characteristics.
Some kind of P2P topology
Logical Addressing of Services and Consumers
I'm not sure that anything else is needed. From here we are just adding bells and whistles. The product that I work on has many more whirly-gigs than I describe here but I think they would only help to loose focus from defining what a Distributed Service Bus is.
See original on c2.com 