Over the past decade or so, many companies have moved their systems to the cloud. But with a plethora of new IoT devices appearing in homes and businesses, there’s a new shift starting to take place in the direction of Edge-based computing.

This means that instead of having computing and memory capacity centralized in the cloud or in a data center, you put it closer to where it’s actually needed. This might mean mobile or handheld devices, intelligent terminals, or even the onboard capabilities of connected vehicles. The advantage of this is that you can deliver results quickly with no need to worry about needing a fast data connection or indeed any connection at all.

Challenges of the Edge

Compared with more traditional computing environments, the Edge presents some unique challenges. The first is that the hardware available is far less powerful. In order to make the best use of it, therefore, it has to be targeted at very specific functions.

Edge systems need to be able to run with little or no intervention too. They may be in remote locations or installed across hundreds of sites, so a low-maintenance system is essential. This usually means some form of remote management capability to allow devices to be monitored and updated as required.

Of course, Edge systems don’t operate in total isolation; they still need to be able to communicate with a data center and this needs to be managed securely in order to reduce the risk of data loss. Edge devices should also have at least some capability to operate independently and cache data in the event of losing a connection to the host or where connections are notoriously unreliable.

Finding the Edge

So, where are we likely to find Edge-based systems appearing? The most obvious example is in retail and hospitality environments, where point of sale terminals can capture and analyse data in order to improve the experience for the customer.

The burgeoning field of Internet of Things (IoT) and Industrial Internet of Things (IIoT) devices is another area in which the Edge is starting to take off. This can mean adding intelligence to heating and ventilation systems, smart energy meters, RFID tracking devices and more. In these instances, data collected from IoT devices and sensors can be processed at the edge to provide for predictive maintenance, demand forecasting, usage tracking and more.

In the near future, we’ll start to see Edge computing used in the development of smart cities, to reduce traffic congestion, provide improved public services and optimize energy use. This is likely to tie into telematics information from connected vehicles to improve navigation and avoid jams. Edge-processed data from vehicles could also be used in setting insurance premiums or improving driver behavior.

As facial recognition technology develops, Edge processing could be used to identify customers and further reduce the risk of fraud in banking and finance businesses.

Closer to the Edge

There is a temptation for technology leaders in companies to rush towards adopting the Edge, as with the cloud and AI, simply because it’s the latest technology. After years of us moving the compute element of the network into the data center, technology manufacturers are now talking about moving it back out, which can be a massive upheaval and expense.

But let’s take a step back from the Edge for a second and think about why you should be considering the use of Edge computing and what benefits it can deliver.

Edge devices can, of course, be combined with another fast-growing technology, artificial intelligence. This allows Edge systems to continuously learn and improve their performance. A good example here would be automated handling systems in warehouses, learning how to arrange stock to best service the supply chain. Edge systems in this kind of environment can also help to minimize the holding of unnecessary inventory and reduce errors. Edge is speed, for example being able to print instant coupons at the checkout.

Edge computing can also benefit business processes. It can be used to streamline just-in-time processes, supply chain interactions and so forth, thanks to the availability of accurate, current data.

But what does all this mean for the business in IT terms? It’s important to look at how it is going to integrate with your existing business systems, whether in the cloud or on-site. What networking technology is needed to achieve this and should you be looking at a combination of virtual machines and containerization? The key consideration is to reduce latency and allow the customer-facing elements of the business to respond quickly to demand.

Most importantly, in any digital transformation strategy, the IT element shouldn’t be a roadblock. Should there be an element of self-provisioning or automation that allows - for example - an increase in computing capacity to cope with peak demands?

Edge devices can, of course, be combined with another fast-growing technology, artificial intelligence. This allows Edge systems to continuously learn and improve their performance. A good example here would be automated handling systems in warehouses, learning how to arrange stock to best service the supply chain. Edge systems in this kind of environment can also help to minimize the holding of unnecessary inventory and reduce errors.

Cautionary notes

Clearly, there are some challenges to effective implementation, not least of which is the extended attack surface presented by, in some cases, relatively simplistic Edge devices with limited security capability. There is little doubt that manufacturers and end user organizations will need to collaborate to enforce strict security and provisioning standards to mitigate this.

Similarly, the hard and soft management of a substantial Edge infrastructure will require careful design and implementation to minimize any impact on already stretched IT and data center staff.

An Edgy future

As the IoT continues to grow, more and more everyday devices from TVs to fridges and industrial plant will gain some degree of computing capability. This means that a lot of the time there will be a good deal of computing power sitting idle. Using this unused Edge power has the potential to create a large and powerful resource, linked by the cloud, that can be used for many purposes.

In many ways, this is an evolution of the existing cloud, but leverages the power of many smaller devices rather than relying upon large data centers. This has the potential to offer far more resilient systems as, rather like the Internet, it becomes possible to reroute and use a different set of resources in the event of failures.

While this vision of the Edge may still be a little way in the future, businesses need to keep a close watch on developments or risk being left behind.