Creating a Digital Ecosystem: Service Orientated Architectures with Distributed Evolutionary Computing

A novel optimisation technique inspired by natural ecosystems is presented where the optimisation works at two levels: a first optimisation, migration of services which are distributed in a decentralised peer-to-peer network operating continuously in time; this process feeds a second optimisation based on a genetic algorithm that operates locally on single peers and is aimed at finding solutions satisfying locally relevant constraints. Through this twofold process, the local search is sped up and yields better local optima as the distributed optimisation already provides a good sampling of the search space by making use of computations already performed in other peers with similar constraints. We call this new distributed optimisation architecture a Digital Ecosystem, which is created by extending a Service Oriented Architecture (SOA) with Distributed Evolutionary Computing (DEC). The EvE Digital Ecosystem will contain services specially supported by application software written to take advantage of the ecosystem. Because of the functionality of the ecosystem these services will recombine and evolve over time, constantly seeking to improve their effectiveness for the user base. This will be achieved through DEC. Individuals within our Digital Ecosystem will be combinations of services (applications), created using evolutionary optimisation, in response to business user requests for applications. These individuals will migrate through the Digital Ecosystem and adapt to find niches where they are useful in fulfilling other business user requests for applications.