A Multi-Agent-Based Data Collection and Aggregation Model for Fog-Enabled Cloud Monitoring

Abstract

The fog-enabled cloud computing has received considerable attention as the fog nodes are deployed at the network edge to provide low latency. It involves various activities, such as configuration management, security management, and data management. Monitoring these activities is essential to improve performance and QoS of fog computing infrastructure. Data collection and aggregation are the basic tasks in the monitoring process, and these phases consume more communicational power as the IoT nodes generate a huge amount of redundant data frequently. In this paper, a multi-agent-based data collection and aggregation model is proposed for monitoring fog infrastructure. The data collection model adopts a hybrid push-pull algorithm that updates the data when a certain change in new data compared to old data. A tree-based data aggregation model is developed to reduce communication overhead between fog node and cloud. The experimental results show that the proposed model improves data coherency and reduces communication overhead compared to existing data collection and aggregation models.