Dynamic iFogSim: A Framework for Full-Stack Simulation of Dynamic Resource Management in IoT Systems

Abstract

Complex Internet of Things (IoT) applications such as Healthcare IoT include a variety of compute, data, and communication kernel intensities and have diverse sensitivities of QoS requirements including latency, throughput, availability, accuracy, etc. Ensuring QoS requirements for the applications requires a comprehensive tool to perform efficient full-stack analysis. Per our observation, the literature lacks a simulator capable of supporting a full-stack communication-computation co-simulation of an IoT system. Furthermore, IoT system behavior can dramatically change during run-time due to variation in status and context. Therefore, such a system must be dynamically controlled over time. In this paper, for the first time, we propose a full-stack framework to co-simulate communication and computation aspects of an IoT system in a dynamic scenario. We integrate a Transmission Control Protocol (TCP) latency model with the iFogSim simulator. We conduct a health-care IoT-based case study to evaluate the framework. The framework is open-sourced and available on GitHub in the following repository: https://github.com/HealthSciTech/Dynamic iFogSim.