Internet of Things and Data Analytics Handbook

42INTERNET OF THINGS AND CLOUD COMPUTING

JAMES OSBORNE

Microsoft, Redmond, WA, USA

42.1 INTRODUCTION

Both “machine‐to‐machine (M2M) systems” and “Internet of Things (IoT)” are broad terms that are sometimes used interchangeably, and there is little point in debating where one ends and the other begins. But it would be a mistake to think of IoT as simply the latest buzzword. It’s more useful to think of IoT as an evolution of M2M made possible by disruptive forces in three key areas of technology (Figure. 42.1):

1. As observed by Moore’s Law [1], integrated circuits have been getting reliably denser (and therefore more powerful) over time. The disruption for IoT was the “jump” of digital technology from fixed devices such as desktop PCs to battery‐operated devices like mobile phones and sensors.
2. Koomey’s Law [2] is less well known, but it’s just as important for IoT because it relates to power consumption. It observes that the number of computations per joule of energy has been doubling about every 1.57 years. Many IoT scenarios involve placing sensors in locations that are not attached to a power grid. In some cases an IoT device may be expected to run on a single battery for 5 years or longer. Because of these types of constraints, IoT developers as a rule must assume that power is not a limitless resource; with Koomey’s Law we can predict that IoT devices will become more and more capable per unit of energy over time and therefore continue to present opportunities for innovation. ...