Fog and Edge Computing

5 Optimization Problems in Fog and Edge Computing

Zoltán Ádám Mann

5.1 Introduction

Fog / edge computing arises through the increasing convergence and integration of several – traditionally distinct – disciplines: cloud computing on one hand, mobile computing and the Internet of Things (IoT) on the other hand, and advanced networking technologies as a glue between them. The main idea is to combine the strengths of these technologies to provide the necessary compute power to end‐user applications in a cost‐effective and secure way, with low latencies. Thus, fog / edge computing brings significant benefits to all of the underlying fields.

The notions of fog computing and edge computing are somewhat vaguely defined in the literature and have largely overlapping meaning [1]. In this chapter, we use the terms “fog computing” and “edge computing” interchangeably to refer to an architecture combining cloud computing with resources on the network edge and end‐user devices.

In cloud computing, there has been an evolution for several years from centralized architectures (one or few large data centers) toward increasing decentralization (several smaller data centers), which is still continuing, and fog / edge computing is a natural next step on this evolution trajectory [2]. Geographically distributed data centers lead to decreased latency for applications involving distributed data sources and sinks (e.g., users or sensors / actuators), since each data source / sink can be served ...