Energy-efficient QoS-aware service allocation for the cloud of things

Cloud computing is a key enabler for the development and deployment of large-scale IoT service platforms. The integration into such platforms of different sensing and actuating systems will imply the availability of many similar IoT services with common functionalities though with different QoS and cost. A cloud-based IoT platform can then benefit from implementing QoS-aware service selection algorithms to match application demand to IoT services, whilst guaranteeing to meet the respective QoS requirements. Such algorithms must however take into account that IoT service providers are usually constrained devices with limited computation, storage and energy capabilities. In this work we formulate the QoS-aware service selection problem for IoT cloud platforms as an integer optimization problem, whose solution minimizes the energy consumption so as to maximize the lifetime of battery-powered devices, whilst guaranteeing the fulfillment of real-time QoS requirements. We then propose a computationally efficient heuristic algorithm to solve the problem, and show through extensive numerical analysis that it is able to find solutions very close to the optimal one in all considered scenarios.