IIT Home Page CNR Home Page

Service Selection and Composition in Opportunistic Networks

Opportunistic computing is a new computational paradigm enabling mobile users to access the heterogeneous services present in a pervasive mobile environment. With respect to conventional service-oriented approaches, in opportunistic computing services are provided by the users' mobile devices themselves, and are accessed exploiting opportunistically direct contacts between devices, i.e. without relying exclusively on fixed infrastructures such as the cloud. Pair-wise contacts are exploited to collect information on services and providers available in the network. A proper support may exploit this information to choose the most efficient composition of services satisfying a service request issued either by a user or an application. This paper defines a support for service selection and composition in opportunistic environments based on a mathematical model able to describe the different phases of the execution of a service composition. The model enables an estimation of the execution time of a composition and is exploited by the support for choosing the best composition among a set of available alternatives. The paper presents a set of simulations proving the effectiveness of our approach. The experiments show that our approach achieves better query resolution time and better load balancing of the service requests on the providers with respect to reference alternative approaches.

The 9th IEEE International Wireless Communications & Mobile Computing Conference – Mobile Computing Symposium (IWCMC 2013), Cagliari, Italy, 2013

IIT authors:

Emanuel Marzini

Foto di Emanuel Marzini

Davide Mascitti

Foto di Davide Mascitti

Type: Article in proceedings of international peer-reviewed conference
Field of reference: Information Technology and Communication Systems