“When you consider a retailer with 2,000 locations across the country, the cost of deploying local data processing and analysis for each location is too high, so edge computing can bring huge benefits,” said Paul Savill, senior vice president of technology The company’s Lumen’s president of product management and services, he pointed out that edge computing is designed to work with the cloud. “Edge nodes combine hardware-driven computing power with software-defined network functions to connect them to the public cloud,” he explained. “From a centralized node in a market area, such as the scale of Denver, edge computing can provide services to more retail locations within five milliseconds.”
Opportunities outweigh challenges
Shivkumar Krishnan, director of Gap store engineering, said that the biggest challenge in implementing edge computing in the retail industry is legacy infrastructure. “As an end user on the cloud, upgrading is much easier because you only need to press a button and shut down or replace the virtual machine. In the retail industry, this is more of a logistics issue,” he explained. When setting up for the first time, each location needs to connect its device to the edge, which may need to be done at night when the customer is not in the store. Suppliers work on site, store security personnel and managers need to be present. Krishnan said: “It really becomes more difficult to figure out everyone’s availability.” “And our 2,500 stores need to repeat this process.” In the cloud, hundreds of servers can be deployed with one click.
When it comes to the Internet of Things and other digital devices, data security is also an inevitable challenge. “The more you concentrate information in one location, the more you need to worry about protecting it, and the greater the risk of creating a single location that can be penetrated and theft of information,” Savill said. However, supporting edge computing on nodes in nearby data centers and connecting to the public cloud is generally safer and more reliable than the retailer’s own. Savill said this is because edge providers, like public cloud providers, provide network security at scale from a central location, so they can understand what threats are and how they affect customers.
In other words, the benefits and opportunities of the edge far outweigh the potential challenges. “One of our biggest edge computing use cases is at the point of sale, where we process millions of transactions,” Krishnan explained. From the store to the cloud, there are many points of failure—switches, routers, telecommunications circuits, and cloud providers. He said: “The edge provides us with a high degree of redundancy, which can handle all transactions in the store itself and return to the cloud when the edge fails.”
“The edge provides us with full redundancy to process all transactions in the store itself, and fall back to the cloud when the edge fails.”
Shivkumar Krishnan, Gap Store Engineering Supervisor
Krishnan said that Gap has invested in edge servers in the past few years as part of the overall platform, using the latest technologies such as microservices, cloud computing, streaming media services and DevOps engineering methods. “Now, with our platform, we can quickly build, verify, and deploy applications-all of which are done in the same day,” he said. “I can remotely monitor and manage most of our more than 100,000 devices. Our sales staff use iPads, which allows us to build intuitive native mobile user experiences.”
Although Gap is the early days of edge computing games, the challenge is to keep up with the latest and most advanced technologies, just like any technology adoption. He said that today’s edge servers have built-in graphics processing units, network routers, and broadband technology 5G, “all of which are encapsulated in small devices built from the ground up for advanced machine learning.” “Hopefully we can seize the next iteration of these advancements and surpass the others who have obtained them now.”