Adaptive Scheduling of Continuous Operators for IoT Edge Analytics

Patient Ntumba, Nikolaos Georgantas, Vassilis Christophides

Abstract: Highlights•Cost Model for evaluating the usage cost of computational and network resources in an Edge–Fog–Cloud (EFC) architecture.•Response Time Model to take into account the constraints related to the time-sensitive nature of data stream processing.•Formulation the problem of adaptive scheduling for DSPA applications across a hierarchical EFC as an optimisation problem.•The objective is to balance the usage of the computational and network resources accross the EFC.•The constraints are related to the available capacities of the two types of resources and the real-time.•Heuristic Algorithm aTSOO-H adaptively schedules DSPA application based on its current deployment state in the EFC architecture.•aTSOO-H algorithm achieves optimal trade-offs between resource usage cost, response time, and execution cost.