A Scheduling Algorithm for a Fog Computing System with Bag-of-Tasks Jobs: Simulation and Performance Evaluation

Cloud computing is evolving in such a way that it is apparent that the future of High Performance Computing (HPC) lies in interconnected heterogeneous systems. Moreover, many of today's needs prefer resources that are diverse in geographic distribution and are close to their location. Hence, Fog Computing was "born" in order to better meet all the demands of today's computational needs. Compared to Cloud Computing, Fog Computing extends it by bringing computing nearer to users, enhancing location-based services and utilizing every available resource. The development and simulation of a Fog Computing System based on Grids, Smartphones, Raspberries and Cloud (Virtual Machines or Containers) is presented in this paper. This work aims at studying the possibilities of using every resource that is available in order to reduce total expenses under a Bag-of-Tasks workload model. Simulation results show that by combining every available resource we can reduce the costs, while the mean response time is not increasing drastically.