Richard Bradley explains Cloud and Edge Computing in Deloitte's broadcast series "Technology Decoded" on World Radio Switzerland.
The physical world – assets, people, places, spaces, supply networks – has historically been ’unconnected’ and therefore largely invisible to business leaders. The Internet of Things (IoT) enables breakthrough results by connecting the physical world to the digital world in innovative ways to enhance or generate new business value.
A conventional view of IoT would portray connected objects transmitting their data through the Cloud to larger data centres where the analytics are conducted and insights are generated. With strained networks and insufficient bandwidth to keep up with real-time demands, Edge computing is an alternative to this process, completing some of these analytics directly with the object’s own processing power.
While IoT solutions have been around for a long time, a decrease in hardware costs and battery consumption, an increase in communication bandwidth and the proliferation of powerful IoT platforms have all paved the way for the IoT to become one of today’s hot technology topics.
As part of our ‘Explained in under 100 words’ series, we’re taking a look at Cloud and Edge Computing and explain how they work – as promised in less than 100 words!
You are a diabetic and use an insulin pump equipped with sensors collecting data from your vital signs.
With Edge computing, the device analyses whether your vitals are within a tolerable range, and triggers responsive actions. Treatment is done in real-time, without human intervention, keeping you healthy, happy and satisfied. As such, monitoring your health status and assigning corresponding treatments is immediate and personalised, raising alarm to emergency contacts before you know your blood sugar is too low.
Edge computing achieves this without reaching out onto the Cloud, reducing the strain on both the healthcare system and network infrastructure.
The IoT’s network-connected objects generate data that is transferred to the Cloud and subsequently stored. This data is used to develop insights and logic, which is then fed back to the objects as required. However, IoT is spreading quickly and straining networks, sending data from objects to hubs, through the network, up to servers – congesting networks with 4G (and now even 5G) technology unable to support the growing load. This data load produces increasing latency and reduces reliability, and these high network connectivity demands produce high infrastructure and network cost.
Edge computing is the logical evolution of the dominant Cloud computing model. Edge computing means pushing computing away from a centralized point, closer to (or even onto) the device, the ‘edge’ or periphery of a network. As such, rather than sending raw data back to a cloud or data center, the device generates action independently or sends only already refined data to the network, in effect storing, processing, analysing and reacting. By keeping the computation and logic on a local basis, this increases the speed and resilience of critical IoT infrastructure, thereby reducing overall network cost and accessibility.
Edge computing also allows for immediate analysis and access to time-based data, also reducing the amount of information subject to potential cyber-attacks. This advantage is also well perceived by firms looking to store their data on-site or to prevent it from being transported across firms, borders and oceans. Furthermore, machine-learning algorithms, which rely on large quantities of recent / relevant data, can access that data without experiencing latency with Edge computing.
Edge computing is one of the central drivers to scale and exploit the Internet of Things, with more and more data from IoT devices getting the ’Edge treatment’. Edge computing is an enabler to the Internet of Things – reducing costs, reaction time and infrastructure needs. Considering that by 2020, the number of IoT devices could exceed 20 billion, and by 2030 there could be more than 500 billion – all up from 8.4 billion as of 2017 – more and more products on the market will need to face up to this new reality.