Edge Computing: What Retail Needs in the IoT Era


July 9, 2019

The world of physical retail has changed a lot. The entire shopping experience has changed, top to bottom. The only way retailers have managed to survive is by changing with the times. One of the things retailers have NOT been changing is the way they compute at their physical location.

A few Point-of-Sale (PoS) systems, a server (or maybe two) and probably some external storage. And quite frankly, that just doesn’t cut the mustard now a days. The Internet of Things (IoT) has totally flipped the way we live (and shop) on our heads.

People want more from their traditional retail experiences and in the IoT era that typically requires a healthy does of digitalization which in turn, will require a healthy amount of computing power (and added security).

In our last article about Edge computing, we mentioned Scale Computing and the work they’re doing with Edge computing devices. They’ve got a pretty good retail plan going and I recommend you check it out (https://www.scalecomputing.com/documents/white-papers/white_papers_retail_on_the_edge.pdf).

Scale’s retail solution focuses on six key things: Resources, Uptime, Virtualization, Built in Data Protection, Store Management and Economics.


With unique, patented software, a Scale Computing HC3 hyper-converged cluster can run on very lean hardware. As for the data handling, the Scale Computing HC3 hyper-converged solution requires less resources than any other hyper-converged solution.

We know that hardware can be expensive. It’s why people found the cloud so promising. If the IoT era has given us anything its small, affordable hardware resources. Edge computing leverages those advances in technology and Scale provides a compact AND affordable option.


A true hyper-converged cluster with built-in replication, cluster-wide redundancy, and automated rolling updates secure the best possible uptime available in the market today.

Uptime can be a huge concern when using Cloud computing. You just don’t have the ability to guarantee that the service you often pay through the nose for will be available to you


Still uncommon for in-store retail IT caused by high reoccurring cost for the visualization software. Scale Computing HC3 hyper-converged solutions comes standard with embedded KVM-based virtualization. There is no additional hypervisor license cost and a new VM can be created quickly and easily, even automated, from the central IT organization down to the store level. The build-in virtualization software is easy to run and manage through the HC3 unified, web-based management console.

Built in Data Protection

The Scale Computing HC3 solution offers integrated replication and failover capabilities and optional off-site cloud protection. Additional, optional security technologies are available such as UPS systems from APC by Schneider or F5 security solutions, running on the hyper-converged cluster as a VM.

Data protection is huge. If big retail hacks (think Target, Home Depot, etc) have taught us anything, it’s that retail is vulnerable. While we’d personally still recommend a security device or two, and a bunch of best security practices on top of ’em, what Scale’s doing with their solution is admirable.

Store Management

With the web-based management interface, thousands of stores can be managed from a single location in a single interface.

Single pane of glass management is an attractive feature in any platform, having the equivalent here is icing on the cake.


The Scale Computing-Lenovo HC3 cluster for retail offers the most economical, expandable, resilient solution available in the market today. Besides a very attractive initial price point, implementation and ongoing management cost are very low. Typically 60-80% lower versus other solutions.

Economics, at least in our opinion, stretches beyond the points mentioned above. Edge computing can also be implemented with zero to no additional real estate cost.

Retail locations are typically tight on space. Every square foot is precious. Rather than spending a lot of money on some out of reach cloud, located miles and miles away, an Edge computing infrastructure can be deployed in a closet or directly mounted to a wall.  

Additional Info

The document I linked to above is good place to start in regards to edge computing. Out side of their six main take aways Scale also focuses a lot on buyer experience in the 21st Century and how that’s impacted retailers.

One of the big things they mention is applications and the need to collect and analyze data from physical locations.

Retail apps typically include things like Point-of-Sale (PoS) apps, pricing apps, scanning apps. These are the run-of-the-mill apps that retail locations have been using for years. But that’s changing too.

Both the traditional apps and the newer, hipper, customer facing applications (like online shopping list tools that will draw you an in-store map to the products you’re looking for) or custom designed furniture tools (like the ones Ikea uses) are incredibly important to the retail experience in the IoT era.

Not only do they enrich the buying experience but they also allow retailers to collect and analyze customer behavior in semi-real time.

That’s what it takes to survive in today’s online-based consumer market. new apps and technologies have to be developed as well. Metrics need to be collected and analyzed constantly in order to survive. The ability to to access that data regularly and with ease is a big part of what makes Edge computing so attractive. Add on top of that the resources Edge computing provides to processing power at a location, enabling all those Apps and services to run natively on site, and you’ve got a real winner on your hands.

*If anybody’s seen the reanimated corpse George A. Romero shuffling around out there looking for a fresh human meal, please apologize for me. The allure of using zombies in a shopping mall was too good not to use.


Carl Keyser is the Content Manager at Integris.

Keep reading

What to Know Before Installing Co-Pilot for Microsoft Word

What to Know Before Installing Co-Pilot for Microsoft Word

Imagine having an AI assistant that pulls from your notes, marries them to an existing document format, and writes a document for you. That's the power of Copilot for Microsoft Word, which is planned for rollout in 2024 for those who buy the Copilot M365 license....

Bridging the Gap between Automation and Innovation

Bridging the Gap between Automation and Innovation

Automation and Innovation. Some people might say those two words cancel each other out. Yet, I believe these two concepts can create capacity for each other—if your business leverages the free time automation creates to foster innovation. Automation can be...