A Framework for QoS Parameters-Based Scheduling for IoT Applications on Fog Environments

Job scheduling in fog computing environments is an evolving area of research and researchers are finding improved ways to assign jobs to the appropriate fog nodes (FN). This paper proposes a job scheduling framework that divides geographical areas into hexagonal regions for better coverage of available resources. Each hexagonal region is considered as a fog environment (FE) and the proposed framework chooses an appropriate FE and FN with the help of brokers. Most of the scheduling frameworks ignore geographic coverage of fog resources, consequently causing the over and under-utilization of available fog resources. The hexagonal shape of FEs in the proposed framework ensures that every single instance of fog resource is covered and utilized for job execution. In addition to this, a Quality of Service (QoS) parameter-based job selection methodology is adopted to choose an FN from the available FEs. Three parameters physical proximity, bandwidth, and Quality of Devices are being considered to calculate a QoS score. The final decision is made based on the comparison of the deadline specified in the Service Level Agreement and the average execution time of the FN. Experimental results of the implementation of the proposed system show better coverage and QoS achievement along with improved resource utilization, resource availability waiting time, response time, and completion time. Experiments are performed on the tasks generated by applications available on the Aneka platform. These tasks are scheduled by a GB (GB), which is a windows 10 machine with Intel(R) Core (TM) i7-10,700 CPU 2.90 GHz processor.