An estimation of distribution algorithm to optimize the utility of task scheduling under fog computing systems

The Internet of Things (IoT) is realized initially today. A large amount of data is produced and a range of IoT services are settled down. Based on it, a range of responsive IoT applications arise. To satisfy the quality of experience (QoE) of users, the applications are needed to be processed in a timely manner. Compared with traditional cloud computing systems, fog computing is one of the promising solutions to processing the huge amount of local data and decreasing the end-to-end latency. Different time-dependent functions are adopted to measure the utility of different tasks and in this work, the resource allocation and task scheduling problem under the fog system is considered to maximize the sum of the utility of tasks. And an estimation of distributed algorithm to maximum the task utility (uEDA) with a repair procedure and local search is adopted to determine the task processing order and computing node allocation. The comparative results show that the performance of our algorithm exceeds significantly the heuristic method on the utility metrics.