Service Evolution in Bio-Inspired Communication Systems

Abstract

An autonomic network must work unsupervised, therefore must be able to respond to unpredictable situations. The BIONETS project is working towards resilient network services that are able not only to perform short-term adapta- tions to the environment but also long-term evolution of new functionalities. To this end, a bio-inspired approach is pro- posed, based on an extension of evolutionary computing to a pervasive environment where disconnected operation is com- mon, and where the fitness of a service is evaluated at runtime. Crossover or recombination of existing services occurs as op- portunistic exchange of parameters or code, producing new generations of services which proliferate in the network or are discarded by a mechanism similar to natural selection. In this paper we review the research lines related to autonomic service evolution currently in progress within BIONETS. A catalytic graph model describes the flow of op- portunistic evolutionary interactions, shaped by cascade fit- ness evaluations. We present a research agenda and possible avenues leading to self-evolving services, and discuss their po- tential impact on future service engineering.