Fog and Edge Computing

9 Predictive Analysis to Support Fog Application Deployment

Antonio Brogi Stefano Forti and Ahmad Ibrahim

9.1 Introduction

Connected devices are changing the way we live and work. In the next years, the Internet of Things (IoT) is expected to bring more and more intelligence around us, being embedded in or interacting with the objects that we will use daily. Self‐driving cars, autonomous domotics systems, energy production plants, agricultural lands, supermarkets, healthcare, and embedded AI will more and more exploit devices and things that are an integral part of the Internet and of our existence without us being aware of them. CISCO foresees 50 billion connected entities (people, machines, and connected things) by 2020 [1], and estimates they will have generated around 600 zettabytes of information by that time, only 10% of which will be useful to some purpose [2]. Furthermore, cloud connection latencies are not adequate to host real‐time tasks such as life‐saving connected devices, augmented reality, or gaming [3]. In such a perspective, the need to provide processing power, storage, and networking capabilities to run IoT applications closer to sensors and actuators has been highlighted by various authors, such as [4, 5].

Fog computing [6] aims at selectively pushing computation closer to where data are produced, by exploiting a geographically distributed multitude of heterogeneous devices (e.g., gateways, micro‐datacenters, embedded servers, personal devices) ...