Service Discovery Framework for Fog Computing

Abstract

Fog computing has experienced significant growth, enabling access to cloud-based services from nearby fog nodes instead of relying on centralized cloud systems. This decentralization allows for the distribution of services across various systems, providing customers with increased proximity and efficiency in accessing the resources they need. In addition to the shared skills of cloud computing, fog computing gives additional features that facilitate mobility, low latency, and real-time interactions. Nevertheless, it remains a tough issue to identify distributed fog computing resources. The purpose of this research was to develop a fog computing service discovery framework that makes use of machine learning. Random Forest, Logistic Regression, GridSearchCV, and Support Vector Machine are only a few of the current machine learning algorithms that the suggested framework uses in the service discovery process. Results are evaluated and contrasted based on the suggested learning model's performance. Because of their specific goals, Random Forest and GridSearchCV are our primary ML algorithms, even if there are others that show superior accuracy.