Deadline and Energy-Aware Application Module Placement in Fog-Cloud Systems

Abstract

Fog computing has emerged as a promising augmentation of cloud computing, positioned at the network's edge, and it is poised to enhance a wide range of Internet of Things (IoT) driven applications. Although fog computing promises to reduce the response time of applications, its omnipresence is subject to the availability and capabilities of the resources in the fog infrastructure. Hence, there is a need of efficiently harness fog infrastructure to execute different IoT applications while meeting their quality of service (QoS) requirements. However, this objective becomes challenging when the applications are decomposed into multiple modules with diverse latency sensitivities. The scatter placement of application modules over distributed fog nodes further intensifies the problem by increasing the overall energy consumption of the fog environment. Therefore, this study proposes a deadline and energy-aware modular application placement policy for fog computing environments. The proposed policy simultaneously prioritizes the placement of critical applications in the fog infrastructure and consolidates the number of active fog nodes for energy management. The performance of the proposed policy was evaluated using iFogSim and compared with several contemporary solutions. The experimental results demonstrate that the proposed policy outperforms others in increasing the percentage of QoS-satisfied applications and reducing energy usage in fog computing.