Blog

Edge Computing: What, Why and How to best Do

Matt Bushell
Matt Bushell
Sr. Director, Product Marketing
July 31, 2019|7 min read

On a typical day, someone might check his FitBit before he hits the gym. Or, he might remotely program his home air conditioner to tick down by five degrees before he gets home. Or, he might check into his hotel room via a smartphone app instead of visiting the front desk to retrieve a key card.

What may have only been seen on “The Jetsons” decades ago is now a reality, and the Internet of Things (IoT) – as seen in smart home apps and personal devices – is changing the way we live our lives and the way businesses serve their customers.

The IoT allows businesses to collect data on customers as never before, but part of the challenge is how to quickly turn that into an advantage. Collecting and analyzing data can take some time, but “edge computing” is a way to not only act on that data more rapidly, but allow the processing of that information to be done closer to where it’s gathered or generated.

When data is collected it can be sent across long routes to data centers or clouds. But with edge computing, that data is collected and analyzed closer to the “edge” of the network. That means that organizations can analyze data in near real-time, which is often critical for many industries such as healthcare, telecommunications or finance.

Edge computing benefits consumers

There are often common questions asked about edge computing, and some of the answers are helping to alleviate the concerns and underscore the benefits. Consider the following:

  1. Is edge computing just another way to collect my personal data? Yes and no. Yes, your personal data is going to be collected, such as your favorite products and where you like to travel. But no, this personal information isn’t just going to be scattered willy-nilly across the universe. Because most mobile devices can now process data effectively without transporting it anywhere, then there is less risk of your data being unnecessarily exposed.

  2. What about security? Because less data is traveling over a network, then edge security can be more secure. In other words, because there’s less data in a corporate data center or cloud, then there’s less chance that it can be hacked by a cybercriminal. On the other hand, security must be top-of-mind when using an edge device in order to stop malicious intent. Data encryption and privacy controls are critical.

  3. Is it really going to make a difference in my life? If you’ve ever glared in frustration at the “spinning circle” that seems to have taken up residence on your smartphone, then you’re going to love edge computing. Edge computing is fast. It greatly reduces latency (the amount of time it takes data to travel) and that means you’re going to be able to see the results you are seeking much, much faster. This will play an even bigger role with emerging 5G wireless networks, which will be able to handle many more connected devices at one time.

  4. Is it reliable? When IoT edge computing devices and edge data centers are closer to customers, then there is less risk of a network problem half a world away slamming into them unexpectedly. It’s much tougher for one failure to shut down a service entirely because data from edge computing devices and data centers can be rerouted through different pathways.

  5. Why should I care about edge computing if I don’t care about IoT? Edge computing isn’t just about tracking your fitness goals or loading your favorite travel app faster. It’s also about the way you do your job. Edge computing is found to be great for helping employees, customers and partners in various locations be able to connect from a centralized location. Then, those users can have access to data and services. It’s a more efficient way to get your job done and work with others.

The business case for edge computing

Last year, HPE CEO Antonio Neri announced the company would be putting $4 billion into edge computing in what is seen as a booming market.

“The opportunity of the edge is about using technology and data to bring personal rewards,” Neri explains. “There is a revolution that’s happening in relation to the explosion of data all around us, being generated at what we call the edge, and that will only continue to grow.”

Consider the benefits of edge computing:

  • Happier customers. Edge computing is able to overcome latency and bandwidth

    problems, which can lead to a better customer experience. For example, let’s say a customer enters a favorite retail store and immediately is offered a coupon for a product that will complement a recent purchase. That interactive, real-time involvement with the customer is seen as a way to compete with online retailers (in 2021, more than 2.14 billion people worldwide are expected to shop online), and to enhance the customer loyalty.

  • More innovation. “Many executives do not understand how edge computing and AI can raise innovation capabilities, operational excellence and customer engagement,”

    says Michelle Duerst, senior director analyst at Gartner. “Edge AI” means that AI techniques are rooted in IoT endpoints, gateways and other devices computing data at the point of use, Gartner reports. By combining edge AI and edge computing with things like augmented and virtual reality or smart factories, businesses can boost innovation.

  • Smarter and faster decisions. Businesses will need to have real-time access to key metrics in order to compete more effectively, implement new strategies and pivot quickly to meet customer demands. Businesses can partner with other organizations to deliver on what the customer needs in the moment, such as a recommendation for a protein smoothie from a nearby shop after working out at a gym, for example.

“The opportunity of the edge is all about using technology and data to bridge the digital and physical worlds,” Neri says. “The edge is where your experiences happen…it is everywhere technology gets put into action, and I personally believe it is the next big opportunity for all of us.”

Pushing infrastructure to the edge

As more companies turn to digital business, it will require that enterprises expand to the edge. Gartner reports that while many companies have a centralized data center or cloud, it’s expected that by 2022 some 75 percent of enterprise-generated data will be created outside those traditional parameters – quite a jump from the less than 10 percent being generated today.

Thomas Bittman, a vice president with Gartner, explains in his analysis of edge computing that edge locations are being created in major metro areas, mobile phone towers, mid-ocean, smartphones and in homes and in cars.

The evolution will require some new thinking and new strategies.

“Edge computing creates challenges for I&O (infrastructure and operations) leaders, including technology choices, distributed computing architectures, remote management and edge security,” Bittman says.

Among the recommendations to meet the challenges, Gartner recommends:

  • Developing an infrastructure strategy that includes edge computing. “Much of the data at the edge will need to be processed there, acted on there and filtered before being forwarded to the core,” Bittman says.

  • Identifying the skills required to handle edge computing, training current workers and hiring new employees when needed. “Teams will need to skill up to address requirements for managing dozens, hundreds or thousands of edge locations that they own or leverage,” Bittman says.

  • Evaluating requirements for privacy, security, regulatory requirements and maturity of public edge offerings as private and shared public capabilities rapidly grow and offer competing choices. “Regulatory requirements may vary by edge location, and security and privacy issues will depend on use cases and even cultures,” Bittman says. “Physical security of edge locations will need to be managed – or planned to be physically unsecure.”

  • Planning for edge market and technology volatility by establishing business cases with rapid ROI. In addition, it should be assumed that some edge deployments may need to be updated or replaced in two to three years.

“As with any fundamentally new IT capability, developing a strategy to leverage edge computing needs to be a shared effort between the I&O organization and the business,” Bittman advises. “Together, they should determine what is technically possible, and how the business can best leverage those technologies.”