A 3.5-tier container-based edge computing architecture
Ching-Han Chen, Chao-Tsu Liu
Abstract
Concomitant with the explosive growth in the number of artificial intelligence Internet of Things (AIoT) devices, a large amount of data is being constantly generated. Further, cloud computing has become increasingly popular for AIoT edge devices. However, challenges such as bandwidth limitations and connection environment constraints exist. To overcome these challenges, distributing computing resources on AIoT gateways or small cloud servers is necessary. In this study, the fog edge computing IoT (FECIoT) architecture was expanded by adding a new hardware layer. Specifically, a 3.5-tier edge computing AIoT (ECAIoT) architecture was developed based on microservices, containers, hardware artificial intelligence engine technology, and an IoT protocol. Experimental results indicate that the request-based load balancing architecture of ECAIoT results in better performance in terms of response time and processing speed. Furthermore. the architecture allows the system to scale flexibly to support different scenarios and demand loads.