The growth of edge computing in retail is undeniable as more devices are connected in stores and more applications rely on the data they create. With the obvious role of the cloud and enterprise data centers in processing transactions and providing data insights, retail IT teams have spent a lot of time determining their overall cloud strategy.
But increasingly, their focus is not just on transactional information across dozens or hundreds of devices in each store, such as inventory, sensors, capacity, video, performance, etc., but on the massive amounts of data processed at the “edge.” I’m here. The role of retail infrastructure in delivering tomorrow’s customer experience has never been more important.
This is both a big challenge and an opportunity for retailers. It’s about how you optimize your edge systems and the data they create, and how you use them as a competitive advantage.
Edge technologies, systems that operate outside the cloud or corporate data centers, form part of large retail architectures that include public clouds, SaaS solutions, e-commerce stacks, distribution centers, and customer-owned or leased data centers. and can all operate “above the store” and connect directly to front-line systems through the network.
In a retail environment, without an edge strategy, a cloud strategy is incomplete at best and doomed to failure at worst.
Many retailers are looking to edge technology as the best approach to modernize their store infrastructure. Everyone agrees that cloud and edge technologies must converge for retail success, but should you take a “cloud out” approach or an “edge in” approach?
I will help you with your decision.
use the right tools for the right job
The cloud, especially the combination of virtualization and containers, improvements in broadband, and heavy investment by hyperscalers (primarily Google, Microsoft, and Amazon) have transformed the way applications run and broken down barriers to innovation. However, like any other tool, the cloud works well for some jobs and poorly for others.
The edge, on the other hand, is designed to handle conditions that cloud infrastructure was not intended to handle. Unlike data centers, edge infrastructures manage unreliable networking, which often causes latency, jitter, and availability issues, especially in large distributed stores.
Store systems typically do not operate in an unattended, highly secure, air-conditioned server environment. Quite the opposite. Edge systems process and control large amounts of checkout and payment data, weigh scales and kitchen systems, manage camera feeds with real-time AI/ML, and best of all, often 24/7. should provide high performance during peak operating hours.
After all, infrastructure designed for the cloud doesn’t fit the edge and vice versa. Cloud systems are typically designed to run in a small number of large, universally available locations and are sometimes referred to as “running at cloud scale.” Conversely, edge systems are designed to run in tens of thousands of small, poorly connected locations and are considered “operating at edge scale.” At a high level, both require scalability and automation, but internally one is highly centralized and the other highly distributed.
For example, retailers need real-time processing such as smooth in-store video recognition, shrinkage detection, product scanning, or generating the latest personalized in-store promotions. The delays and costs caused by high bandwidth demands create friction that impacts both customer experience and revenue. Edge systems process data in real time. However, analyzing this data over time requires the power of a true offline data center/cloud optimized by cloud systems.
With so many moving parts and distributed stores, your edge infrastructure must be responsive, scalable, and reliable. “Edge-in” is the best approach for stores. In contrast, systems sitting on top of headquarters, store data centers, or e-commerce are more cloud-like and better suited for a “cloud-out” strategy.
don’t think either way
In practice, we need to design solutions that work best at the edge of the store while taking full advantage of the cloud. The cloud is integral to the design, but not alone, as it is ideal for managing offline data analytics, building learning models, processing big data, e-commerce, and more.
If you want to deliver real-time promotions to in-store consumers, host your application on the edge to respond instantly and train artificial intelligence (AI) algorithms in the cloud to push model changes to the edge.
Additionally, the use of the term cloud should be clarified. Take cloud point of sale (POS) as an example. POS is one of the biggest use cases for edge computing. With millions of transactions processed, there are many potential points of failure, including switches, routers, networks, and cloud providers.
If the POS is a true cloud application (a web browser connected to the cloud over the internet) and the connection is lost, there is no POS in the store. Cloud POS needs to be edge-ready to keep working. So, in effect, it’s a hybrid solution that includes both cloud and edge.
Streamline your approach
If crowd-out and edge-in methods are not mutually exclusive, a holistic approach is required to maximize the benefits of both. Only then can it be cost effective.
Why stream all your data 24/7 to the cloud when you have gateway capabilities that give you the flexibility to choose which data goes to the cloud and which runs at the edge? . A hybrid approach simplifies operations.
Focus on business-based outcomes
Rather than competing with cloud computing, edge computing complements and completes each other. For retail, the edge is in some ways just a distributed cloud, but a completely different type of cloud.
Cloud computing has provided efficient technology resources and centralized applications on a pay-as-you-go basis. Hyperscalers are slowly finding ways to push the most relevant pieces of their technology from those data centers to other clouds and data centers. But be careful about deploying reused infrastructure from the cloud to the edge.
For example, don’t ask the engineering team that built a large nuclear power plant to use the same approach to build thousands of distributed wind turbines to generate energy hundreds of miles offshore. I guess. Edge needs tools designed for a different job.
The edge represents an opportunity to consolidate old infrastructure for greater efficiency and embrace new high-data, low-latency use cases to improve customer experience. You need to converge the cloud and the edge. Doing so makes it easier to deploy and manage applications where they are best, using consistent policies and configurations across architectures, without requiring lengthy development cycles.
As retailers navigate this transition, they need to focus on business benefits rather than trying to solve specific technology problems. By taking this approach, you can quantify business outcome-based solutions with specific service level agreements (SLAs).
In summary, the question is not “crowd out” versus “edge in”, but how the two approaches can be combined to maximize store operations and drive success. By following these best practices and leveraging the expertise of the right provider, you can quickly answer this question.
One final thought. Retailers are demanding the same agility for in-store applications as they are for digital channels. The edge is the key to enabling the technology that makes it happen.
Nick East is vice president, NCRMore, leading the retail edge software team to transform edge computing for retailers to maximize the performance of their store infrastructure. Prior to joining NCR, he was co-founder and CEO of Zynstra. Zynstra is an award-winning leader in dedicated retail edge software, which he sold to NCR for $130 million. East was an early employee of telecommunications industry software leader Cramer, where he served as GM of the OSS division from its inception until his $425 million acquisition by Amdocs.
