3Edge Computing
Mohammad Hossein Zoualfaghari1, Simon Beddus1, and Salman Taherizadeh2
1British Telecommunications plc, Ipswich, UK
2Jožef Stefan Institute, Ljubljana, Slovenia
3.1 Introduction
Today, hyper‐scale cloud computing [1] from the likes of Amazon, Google, and Azure offers a very cost‐effective solution for storage, processing, and analysis of Internet of Things (IoT) data. The economics of cloud provision relies on a limited number of huge datacenters that are normally located many hundreds of miles away from IoT devices. This model of operations works well for many applications such as web browsing and email, which are tolerant of network packet loss, latency, and jitter typical of cloud computing.
For those applications that need predictable network reliability, security, and low‐latency data processing, the edge computing model [2] offers an answer. Edge computing provides data processing geographically close to assets such as sensors, actuators, IoT objects, and humans. Colocation reduces distances, latency, and in many cases solution complexity that results in a better outcome for customers and end‐users. Edge video analytics processing can eliminate the need to stream video data to cloud‐based service, thus reducing the network load.
Since edge resources are typically in a private network, customers may also choose edge computing due to factors such as having added control of data privacy and the reliability of mission critical applications. Another benefit of ...
Get The Internet of Things now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.