A survey on nature-inspired techniques for computation offloading and service placement in emerging edge technologies

Abstract

Internet of Things (IoT) aims to make an environment more innovative and productive by connecting physical things to the internet. Processing generated data from IoT devices and actuation warranted in real-time requires computational infrastructure near the edge to get the outcome without delay. Emerging edge technologies such as Fog computing, Multi-Access Edge Computing, and Cloudlet provide computing resources near the edge, i.e. closer to the IoT devices, where devices can place their services/applications or offload their computational job for processing. The utilization of computing resources provided by emerging edge technologies addresses the issue of delay in the outcome and increases the battery life of IoT devices/End-user devices. Computational resources provided by the edge technologies, i.e. edge/fog nodes, can be heterogeneous, dynamic and mobile. Therefore, service placement and computation offloading on edge/fog nodes are challenging issues, and the problem to finding the best suitable fog/edge nodes is NP-Hard. Nature-inspired algorithms provide robust solutions to NP-Hard problems. Nowadays, nature-inspired algorithms have been widely applied for resource allocation for service placement and computation offloading in emerging edge technologies. In this work, we provide a detailed study on the applications of nature-inspired algorithms in emerging edge computing domains. We provide an overview of emerging edge technologies, related quality parameters and nature-inspired algorithms followed by the basic formulation of service placement and computation offloading in emerging edge computing systems. In this work, we classify the works in emerging edge computing applying nature-inspired algorithms into two categories: works related to service placement and works related to offloading. We provide a thorough review and comparison of the existing nature-inspired approaches in each category. We discuss various open issues at the end to set future research directions. © 2022, The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature.