Latency-aware Internet of Things Scheduling in Heterogeneous Fog-Cloud Paradigm

Nowadays, with the advent of many new technologies, the data is alarmingly generated by the widespread of internet devices in this data world. Cloud computing seems a viable option for scheduling dynamic data with disparate specifications. However, the execution time increases due to the computationally-limited resources causing latency overhead. So, Fog computing has evolved as a promising paradigm to complement Cloud computing. Therefore, effectively utilizing the underlying resources for scheduling enormous tasks generated by the latency-sensitive applications is a critical issue. Hence, to cope with this, the current research considers a Multi-Level Feedback Queue (MLFQ) for task classification depending on the priority of each layer to reduce the latency and waiting time. Moreover, the dynamic tasks are scheduled using a heuristic-based approach. A proposed objective function is optimized through the heuristic-based method for the minimization of latency rate, makespan, and maximization of resource utilization. Dynamic tasks and heterogeneous resources in the Fog-Cloud environment are considered for appraising the nature of heterogeneity. Extensive simulations are carried out and the obtained results demonstrate the novelty of the proposed algorithm against state-of-the-art methods for conflicting scheduling criteria.