Performance Modeling and Suitability Assessment of Data Center Based on Fog Computing in Smart Systems

In human-centered smart systems, traditional cloud computing has been well adopted for several years for supporting IoT applications of smart systems, however, the requirements of the IoT applications vary due to different tasks in the smart systems. In order to offer location-aware, latency-sensitive monitoring, and intelligent control for IoT applications in the smart systems like smart home or smart grid, we introduce fog computing and discuss the combination of fog computing and data center hoping to improve the quality of experience of users in human-centered smart systems. In this paper, we propose a three-layer architecture of the data center by introducing the fog computing layer between the traditional centralized data center and terminal nodes. Performance related service latency of different cloud services is modeled and evaluated based on it. It is verified that the new architecture has a great improvement in terms of service latency and network load by simulation experiments.