It is getting increasingly recognized that the models and tools of standard service-oriented architectures are not adequate to tackle the decentralized, pervasive, and very dynamic scenarios of modern ICT (Information and Communication Technologies) systems, and that innovative and flexible software architectures have to be identified. This paper discusses how these architectures could get inspiration from natural systems, so as to enforce those features of self-adaptability and evolvability that are inherent in natural systems. In particular, we propose to get inspiration from ecological systems to model and deploy services as autonomous individuals, spatially-situated in an ecosystem of other services, data sources and pervasive devices. Services will be able to self-organize their interaction patterns according to a sort of "food web" and in respect of a limited set of interaction laws. Accordingly, the paper introduces a general reference architecture to frame the key concepts of our ecological approach, details its characteristics, and also with the help of a case study, discusses its implementation and presents simulation results to show the effectiveness of the approach. © 2010 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Villalba, C., Mamei, M., & Zambonelli, F. (2010). A self-organizing architecture for pervasive ecosystems. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6090 LNCS, pp. 275–300). https://doi.org/10.1007/978-3-642-14412-7_13
Mendeley helps you to discover research relevant for your work.