Distributed Workload Control for Federated Service Discovery

Abstract The diffusion of the internet paradigm in each aspect of human life continuously fosters the widespread of new technologies and related services. In the Future Internet scenario, where 5G telecommunication facilities will interact with the internet of things world, analyzing in real time big amounts of data to feed a potential infinite set of services belonging to different administrative domains, the role of a federated service discovery will become crucial. In this paper the authors propose a distributed workload control algorithm to handle efficiently the service discovery requests, with the aim of minimizing the overall latencies experienced by the requesting user agents. The authors propose an algorithm based on the Wardrop equilibrium, which is a game-theoretical concept, applied to the federated service discovery domain. The proposed solution has been implemented and its performance has been assessed adopting different network topologies and metrics. An open source simulation environment has been created allowing other researchers to test the proposed solution.