Joint Optimization of Fog Server Workload and Task Completion Time in Fog Computing

Fog computing as a promising technology develops network service quality by bringing resources closer to the end users. Unlike traditional cloud, fog is characterized by sporadic resources availability, geographical distribution and increased flexibility. However, tasks offloading and scheduling mechanisms still meet challenges in fog server workload balancing and task completion time minimizing. For solving these problems, a task scheduling strategy based on a general IoT-Fog analysis model is proposed in this paper. In the strategy, we will study how the end device selects the suitable fog server to process the requests. Based on the strategy, the end devices offload their tasks to specific fog server based on the propagation latency, computation capability of the server, etc. Without loss of generality, we consider the problem in probabilistic manner, such as request generating rate of the end device and the service rate of the fog server. Finally, extensive simulation experiments are done to demonstrate the effectiveness of our proposed strategy.