Transaction Throughput Maximization under Delay and Energy Constraints in Fog-IoT Networks

Abstract

In this paper, we consider a Fog-IoT network, comprising multiple terminal nodes (TNs) as well as fog nodes (FNs), in which each TN first determines the proper FN to be associated with different quality of service (QoS) and quality of transmission (QoT) constraints, such as energy efficiency, processing delay, and bandwidth requirements; then, it sends its tasks to the associated FN. The main objective of this paper is to maximize the transaction throughput of FNs, i.e., the number of tasks that FNs can process, while there exists a throughput fairness as well as an energy consumption fairness among all FNs in the network. To this end, two algorithms, namely TF-TRADE and QL-TRADE, are proposed. The TF-TRADE aims to ensure throughput fairness and at the same time optimize the total transaction throughput of the network. The QL-TRADE not only follows the same objective, but also tries to balance the energy consumption among all FNs. Analysis and simulation results reveal that the proposed algorithms could alleviate the network performance in terms of the transaction throughput as well as the energy consumption balance among all FNs.