Multiple linear regression-based energy-aware resource allocation in the Fog computing environment

Fog computing is a promising computing paradigm for processing time-sensitive Internet of Things (IoT) applications. It helps process application requests close to the users to deliver faster processing outcomes than the Cloud by minimising overall response time. The Fog computing computation environment is highly dynamic regarding resource availability and communication. Furthermore, most Fog devices are battery-powered; hence, the chances of the failure of the application processing is high, leading to delaying the application outcome. If we process the application requests on other available devices after the failure occurs, it might cause delay and may not comply with time-sensitive requirements. To avoid application processing failure due to power unavailability, we can run applications in an energy-aware manner. This is a challenging task due to the dynamic nature of the Fog computing environment. It is required to allocate resources for application requests so that the application processing should not fail due to the unavailability of power. In this paper, we propose a multiple linear regression-based resource allocation mechanism to run applications with energy-awareness in the Fog computing environment. This approach minimises failures due to power constraints of the devices. Prior works lack energy-aware application execution considering the dynamism of the Fog computing environment. Hence, we propose multiple linear regression-based approaches to achieve energy-awareness objectives. We present a sustainable energy-aware framework and algorithm that executes applications in a Fog environment in an energy-aware manner that minimises application execution failures. The trade-off between energy-efficient allocation and application execution time has been investigated and shown to have a minimum negative impact on the system while employing energy-aware allocation. The evaluation of the proposed method is carried out in a controlled simulation environment by extending CloudSim toolkit. We compared our proposed method with existing approaches. Our proposed approach minimises the delay and processing time by 20%, and 17% compared with the existing one. Furthermore, SLA violations decreased by 57% for the proposed energy-aware allocation.