A novel multi-objective optimized DAG task scheduling strategy for fog computing based on container migration mechanism

Abstract

In order to solve the complex task scheduling problem of fog computing processing big data in the industrial Internet of Things, a task scheduling strategy based on ant colony algorithm called TSSAC (task scheduling strategy with ant colony)is proposed. Tasks with dependencies are modeled as a directed acyclic graph. The performance indices including makespan, load balancing and energy consumption of fog server are optimized simultaneously, and the ant colony algorithm is used to solve the multi-objective optimization problem. The pheromone heuristic factor and pheromone evaporation coefficient of the ant colony algorithm are updated in a linear increasing way, so that the ants are less affected by pheromones in the early stage and obtain a larger search range. During the later stage, it is greatly affected by pheromone and quickly converges to the optimal solution. Furthermore, during the task execution container migration mechanism is introduced to solve the overload problem caused by high server utilization and energy loss caused by low server utilization simultaneously. The simulation results show that the proposed task scheduling strategy TSSAC reduces energy consumption by 23.5% compared with the traditional algorithm, meanwhile, achieves a compromise between task makespan and load balancing index compared with the traditional algorithm.